From 290f5515b71edb1a7f342c7a298a8b0c94de34b9 Mon Sep 17 00:00:00 2001
From: Krazerleo <krazerleo@gmail.com>
Date: Mon, 25 Apr 2022 16:23:25 +0700
Subject: [PATCH 01/17] update scripts

---
 Assets/Scripts/Bots/CharacterFactory.cs       | 18 +++++----
 .../Scripts/Character/CharacterCondition.cs   | 17 +++++++--
 .../Scripts/Character/MovementController.cs   | 21 ++++++++--
 Assets/Scripts/Character/NPC.cs               | 27 ++++++++-----
 Assets/Scripts/Managers/GameManager.cs        | 14 +++++++
 Assets/Scripts/Misc/FlagZone.cs               |  6 +--
 Assets/Scripts/Misc/Settings.cs               | 38 ++++++++++---------
 Assets/Scripts/Pickups/AmmoPickUp.cs          |  2 +-
 Assets/Scripts/Pickups/ArmourPickUp.cs        |  2 +-
 Assets/Scripts/Pickups/HealthPickUp.cs        |  2 +-
 Assets/Scripts/Pickups/PickUpSpawner.cs       |  2 +-
 Assets/Scripts/Utils/BoolToInteger.cs         |  7 ++++
 Assets/Scripts/Utils/BoolToInteger.cs.meta    | 11 ++++++
 13 files changed, 120 insertions(+), 47 deletions(-)
 create mode 100644 Assets/Scripts/Utils/BoolToInteger.cs
 create mode 100644 Assets/Scripts/Utils/BoolToInteger.cs.meta

diff --git a/Assets/Scripts/Bots/CharacterFactory.cs b/Assets/Scripts/Bots/CharacterFactory.cs
index e2c8be3..ab18289 100644
--- a/Assets/Scripts/Bots/CharacterFactory.cs
+++ b/Assets/Scripts/Bots/CharacterFactory.cs
@@ -25,10 +25,10 @@ public class CharacterFactory : MonoBehaviour
 
     private void Start()
     {
-        var attcNum = SettingsReader.Instance.GetSettings.numOfAttackers;
-        var defNum = SettingsReader.Instance.GetSettings.numOfDefenders;
-        var humanDef = SettingsReader.Instance.GetSettings.hasHumanDefender == true ? 1 : 0;
-        var humanAtc = SettingsReader.Instance.GetSettings.hasHumanAttacker == true ? 1 : 0;
+        var attcNum = SettingsReader.Instance.GetSettings.NumOfAttackers;
+        var defNum = SettingsReader.Instance.GetSettings.NumOfDefenders;
+        var humanDef = SettingsReader.Instance.GetSettings.HasHumanDefender == true ? 1 : 0;
+        var humanAtc = SettingsReader.Instance.GetSettings.HasHumanAttacker == true ? 1 : 0;
 
         if (humanAtc == 1 && humanDef == 1)
             throw new System.ArgumentException("Can be only one human player");
@@ -56,6 +56,10 @@ public class CharacterFactory : MonoBehaviour
             spawnPoint.position,
             Quaternion.identity);
         gameobject.SetActive(true);
+        if (team == Team.Attackers)
+            gameObject.tag = "Attacker";
+        else
+            gameObject.tag = "Defender";
 
         if (typeAi == TypeAI.HumanAI)
         {
@@ -65,7 +69,7 @@ public class CharacterFactory : MonoBehaviour
         else
         {
             gameobject.GetComponent<NPC>().GetCharacter.Team = team;
-            gameobject.GetComponent<MovementController>().currentPosition = spawnPoint;
+            gameobject.GetComponent<MovementController>().CurrentNavPoint = spawnPoint;
             Bots.Add(gameobject);
         }
     }
@@ -81,8 +85,8 @@ public class CharacterFactory : MonoBehaviour
             else
                 bot.transform.position = spawnPointsForDefendersTeam[Random.Range(0, spawnPointsForDefendersTeam.Count)].position;
         }
-        var player = Player.GetComponent<Player>();
-        if (player != null)
+        Player player;
+        if (TryGetComponent<Player>(out player))
         {
             player.ResetCharacter();
             if (player.GetCharacter.Team == Team.Attackers)
diff --git a/Assets/Scripts/Character/CharacterCondition.cs b/Assets/Scripts/Character/CharacterCondition.cs
index d1ec569..b674222 100755
--- a/Assets/Scripts/Character/CharacterCondition.cs
+++ b/Assets/Scripts/Character/CharacterCondition.cs
@@ -21,6 +21,17 @@ public class CharacterCondition
             OnChangeHealthEvent?.Invoke(value);
         } 
     }
+
+    public int GetHealthPointsInQuantile()
+    {
+        if (health < 25)
+            return 0;
+        else if (health < 50)
+            return 1;
+        else if (health < 75)
+            return 2;
+        else return 3;     
+    }
     private int armour;
     public int ArmourPoints
     {
@@ -51,9 +62,9 @@ public class CharacterCondition
     public CharacterCondition()
     {
         var settings = SettingsReader.Instance.GetSettings;
-        ammo = settings.maxAmmo;
-        health = settings.maxHealth;
-        armour = settings.maxArmour;
+        ammo = settings.MaxAmmo;
+        health = settings.MaxHealth;
+        armour = settings.MaxArmour;
     }
 
     public void GiveHealth(int health) => HealthPoints = Mathf.Clamp(health + HealthPoints, 0, 100);
diff --git a/Assets/Scripts/Character/MovementController.cs b/Assets/Scripts/Character/MovementController.cs
index 20e1026..334b7d2 100644
--- a/Assets/Scripts/Character/MovementController.cs
+++ b/Assets/Scripts/Character/MovementController.cs
@@ -2,16 +2,31 @@
 using System.Collections.Generic;
 using UnityEngine;
 using UnityEngine.AI;
+using System.Threading.Tasks;
 
 [RequireComponent(typeof(NavMeshAgent))]
 public class MovementController : MonoBehaviour
 {
-    public NavPoint currentPosition { get; set; }
+    public NavPoint CurrentNavPoint { get; set; }
+    public float FlagDistance { get; private set; }
+    private GameObject flag;
+    private const float updateFlagPositionDelay = 5; 
     [SerializeField] private NavMeshAgent navMeshAgent;
 
     private void Start()
     {
-        navMeshAgent.speed = SettingsReader.Instance.GetSettings.movementSpeed;       
+        navMeshAgent.speed = SettingsReader.Instance.GetSettings.MovementSpeed;
+        InvokeRepeating(nameof(UpdateFlagPosition), 0, updateFlagPositionDelay);
+    }
+
+    private void OnDestroy()
+    {
+        CancelInvoke(nameof(UpdateFlagPosition));
+    }
+
+    private void UpdateFlagPosition()
+    {
+        FlagDistance = (flag.transform.position - gameObject.transform.position).magnitude;
     }
 
     public void MoveToRandomPoint()
@@ -23,7 +38,7 @@ public class MovementController : MonoBehaviour
     public List<NavPoint> getPointsCandidate()
     {
         return MapManager.navPoints
-            .Where(point => (currentPosition.position - point.position).magnitude < SettingsReader.Instance.GetSettings.movementSpeed)
+            .Where(point => (CurrentNavPoint.position - point.position).magnitude < SettingsReader.Instance.GetSettings.MovementSpeed)
             .ToList();
     }
 
diff --git a/Assets/Scripts/Character/NPC.cs b/Assets/Scripts/Character/NPC.cs
index 1069d42..c738d0b 100644
--- a/Assets/Scripts/Character/NPC.cs
+++ b/Assets/Scripts/Character/NPC.cs
@@ -3,7 +3,6 @@ using UnityEngine;
 using Unity.MLAgents;
 using Unity.MLAgents.Sensors;
 using Unity.MLAgents.Actuators;
-using System.Collections.Generic;
 
 [RequireComponent(typeof(MovementController))]
 public class NPC : Agent, ICharacter
@@ -11,6 +10,7 @@ public class NPC : Agent, ICharacter
     [HideInInspector]
     public Character AgentCharacter;
     public CharacterCondition Condition;
+    private FlagZone flagZone;
 
     public NPC_BaseState NPC_State { get; private set; }
 
@@ -47,29 +47,38 @@ public class NPC : Agent, ICharacter
     public override void OnEpisodeBegin()
     {     
         NPC_State = DirectState;
+        flagZone = GameObject.FindObjectOfType<FlagZone>();
     }
 
     public override void CollectObservations(VectorSensor sensor)
     {
+        var candidates = moveController.getPointsCandidate();
+
         sensor.AddObservation(Condition.HealthPoints);
         sensor.AddObservation(Condition.ArmourPoints);
         sensor.AddObservation(Condition.Ammunition);
         sensor.AddObservation((int)NPC_State.State);
-
-        var candidates = moveController.getPointsCandidate();
+        sensor.AddObservation((!flagZone.isNotOccup).ToInt());
+        sensor.AddObservation(AgentCharacter.LastTimeHit);
+        sensor.AddObservation(Condition.GetHealthPointsInQuantile());
+        sensor.AddObservation(candidates.Count);
+        sensor.AddObservation(GameManager.IsEnemyNearby(gameObject.transform.position, AgentCharacter.Team));
+        
         foreach (var point in candidates)
         {
+            Debug.Log((float)moveController.CurrentNavPoint.PointId);
+
             bufferSensor.AppendObservation(new float[] {
                 //1 position in navpointId
-                (float)moveController.currentPosition.PointId,
+                (float)moveController.CurrentNavPoint.PointId,
                 //2 distance to flag
-                moveController.currentPosition.FlagDistance,
+                moveController.FlagDistance,
                 //3 death count in point
-                moveController.currentPosition.DeathAttr,
+                moveController.CurrentNavPoint.DeathAttr,
                 //4 flagEnemyDistance
-                GameManager.IsCloserToFlagFromNextNavPoint(point, transform.position)==true?1:0,
+                GameManager.IsCloserToFlagFromNextNavPoint(point, transform.position).ToInt(),
                 //5 EnemyVsNavPointDistance
-                GameManager.IsCloserToEnemyThanToNextNavPoint(point,transform.position, AgentCharacter.Team)==true?1:0
+                GameManager.IsCloserToEnemyThanToNextNavPoint(point,transform.position, AgentCharacter.Team).ToInt()
             }); 
 
         }
@@ -103,7 +112,7 @@ public class NPC : Agent, ICharacter
         if (Condition.HealthPoints < 0)
         {
             OnKilledEvent?.Invoke(this);
-            moveController.currentPosition.DeathAttr += 1;
+            moveController.CurrentNavPoint.DeathAttr += 1;
         }
     }
 
diff --git a/Assets/Scripts/Managers/GameManager.cs b/Assets/Scripts/Managers/GameManager.cs
index 54d1e1d..b91908d 100755
--- a/Assets/Scripts/Managers/GameManager.cs
+++ b/Assets/Scripts/Managers/GameManager.cs
@@ -51,6 +51,20 @@ public class GameManager : MonoBehaviour
         return false;
     }
 
+    public static bool IsEnemyNearby(Vector3 currentTransform, Team team)
+    {
+        SimpleMultiAgentGroup agentGroup;
+        if (team == Team.Attackers)
+            agentGroup = AttackersTeam;
+        else
+            agentGroup = DefendersTeam;
+
+        foreach (var agent in agentGroup.GetRegisteredAgents())
+            if ((currentTransform - agent.transform.position).magnitude < SettingsReader.Instance.GetSettings.ViewDistance)
+                return true;
+        return false;
+    }
+
     public static bool IsCloserToFlagFromNextNavPoint(NavPoint navPoint, Vector3 currentTransform)
         => navPoint.FlagDistance < (currentTransform - GameObject.FindGameObjectWithTag("Flag").transform.position).magnitude;
 
diff --git a/Assets/Scripts/Misc/FlagZone.cs b/Assets/Scripts/Misc/FlagZone.cs
index 8cd4ab4..c6328dd 100755
--- a/Assets/Scripts/Misc/FlagZone.cs
+++ b/Assets/Scripts/Misc/FlagZone.cs
@@ -9,14 +9,14 @@ public class FlagZone : MonoBehaviour
     public float TimeStayDefenders { get; private set; }
     private int occupDefenders;
     private int occupAttackers;
-    private bool isOccupBoth => (occupDefenders>0) && (occupAttackers>0);
-    private bool isNotOccup => (occupDefenders == 0) && (occupAttackers == 0);
+    public bool isOccupBoth => (occupDefenders>0) && (occupAttackers>0);
+    public bool isNotOccup => (occupDefenders == 0) && (occupAttackers == 0);
     private float timeForWin;
 
     private void Start()
     {
 
-        timeForWin = SettingsReader.Instance.GetSettings.timeToWin;
+        timeForWin = SettingsReader.Instance.GetSettings.TimeToWin;
         TimeStayAttackers = 0;
         TimeStayDefenders = 0;
         occupAttackers = 0;
diff --git a/Assets/Scripts/Misc/Settings.cs b/Assets/Scripts/Misc/Settings.cs
index e01abc1..b1b7ea8 100755
--- a/Assets/Scripts/Misc/Settings.cs
+++ b/Assets/Scripts/Misc/Settings.cs
@@ -3,30 +3,32 @@
 [CreateAssetMenu(fileName ="Game Settings", menuName = "Game/Settings", order = 51)]
 public class Settings : ScriptableObject
 {
-    public bool isTesting;
+    public bool IsTesting;
 
-    public float timeToWin;
-    public float timeOut;
+    public float TimeToWin;
+    public float TimeOut;
 
     [Header("movement")]
-    public float movementDistance;
-    public float movementSpeed;
+    public float MovementDistance;
+    public float MovementSpeed;
 
-    public TypeAI defTeamAI;
-    public TypeAI atcTeamAI;
-    public int numOfDefenders;
-    public int numOfAttackers;
-    public bool hasHumanDefender;
-    public bool hasHumanAttacker;
+    public TypeAI DefTeamAI;
+    public TypeAI AtcTeamAI;
+    public int NumOfDefenders;
+    public int NumOfAttackers;
+    public bool HasHumanDefender;
+    public bool HasHumanAttacker;
 
-    public int healthPickupAmount;
-    public int armourPickupAmount;
-    public int ammunitionPickupAmount;
-    public int pickupsAmount;
+    public int HealthPickupAmount;
+    public int ArmourPickupAmount;
+    public int AmmunitionPickupAmount;
+    public int PickupsAmount;
 
-    public int maxHealth;
-    public int maxArmour;
-    public int maxAmmo;
+    public int MaxHealth;
+    public int MaxArmour;
+    public int MaxAmmo;
+
+    public float ViewDistance;
 
     public float GetHitChanceInDirectPoint;
     public float GetHitChanceInRunning;
diff --git a/Assets/Scripts/Pickups/AmmoPickUp.cs b/Assets/Scripts/Pickups/AmmoPickUp.cs
index 506598d..8439c21 100755
--- a/Assets/Scripts/Pickups/AmmoPickUp.cs
+++ b/Assets/Scripts/Pickups/AmmoPickUp.cs
@@ -13,7 +13,7 @@ public class AmmoPickUp : MonoBehaviour, IPickable
 
     public void PickObject(GameObject obj)
     {
-        obj.GetComponent<ICharacter>()?.GetCharacter.Condition.TakeAmmo(SettingsReader.Instance.GetSettings.ammunitionPickupAmount);
+        obj.GetComponent<ICharacter>()?.GetCharacter.Condition.TakeAmmo(SettingsReader.Instance.GetSettings.AmmunitionPickupAmount);
         gameObject.SetActive(false);
     }
 }
diff --git a/Assets/Scripts/Pickups/ArmourPickUp.cs b/Assets/Scripts/Pickups/ArmourPickUp.cs
index 86be289..c68e491 100755
--- a/Assets/Scripts/Pickups/ArmourPickUp.cs
+++ b/Assets/Scripts/Pickups/ArmourPickUp.cs
@@ -13,7 +13,7 @@ public class ArmourPickUp : MonoBehaviour, IPickable
 
     public void PickObject(GameObject obj)
     {
-        obj.GetComponent<ICharacter>()?.GetCharacter.Condition.GiveArmour(SettingsReader.Instance.GetSettings.armourPickupAmount);
+        obj.GetComponent<ICharacter>()?.GetCharacter.Condition.GiveArmour(SettingsReader.Instance.GetSettings.ArmourPickupAmount);
         gameObject.SetActive(false);
     }
 }
diff --git a/Assets/Scripts/Pickups/HealthPickUp.cs b/Assets/Scripts/Pickups/HealthPickUp.cs
index f92345a..36f81ba 100755
--- a/Assets/Scripts/Pickups/HealthPickUp.cs
+++ b/Assets/Scripts/Pickups/HealthPickUp.cs
@@ -13,7 +13,7 @@ public class HealthPickUp : MonoBehaviour, IPickable
 
     public void PickObject(GameObject obj)
     {
-        obj.GetComponent<ICharacter>()?.GetCharacter.Condition.GiveHealth(SettingsReader.Instance.GetSettings.healthPickupAmount);
+        obj.GetComponent<ICharacter>()?.GetCharacter.Condition.GiveHealth(SettingsReader.Instance.GetSettings.HealthPickupAmount);
         gameObject.SetActive(false);
     }
 }
diff --git a/Assets/Scripts/Pickups/PickUpSpawner.cs b/Assets/Scripts/Pickups/PickUpSpawner.cs
index 945c1d3..7a29705 100755
--- a/Assets/Scripts/Pickups/PickUpSpawner.cs
+++ b/Assets/Scripts/Pickups/PickUpSpawner.cs
@@ -27,7 +27,7 @@ public class PickUpSpawner : MonoBehaviour
     private void Start()
     {
         pickups = new List<GameObject>();
-        var amount = SettingsReader.Instance.GetSettings.pickupsAmount;
+        var amount = SettingsReader.Instance.GetSettings.PickupsAmount;
         for (int i = 0; i < amount; i++)
             pickups.Add(GameObject.Instantiate(healthPrefab, spawnPoints[Random.Range(0, spawnPoints.Count)].transform.position, Quaternion.identity));
         for (int i = 0; i < amount; i++)
diff --git a/Assets/Scripts/Utils/BoolToInteger.cs b/Assets/Scripts/Utils/BoolToInteger.cs
new file mode 100644
index 0000000..0d8090a
--- /dev/null
+++ b/Assets/Scripts/Utils/BoolToInteger.cs
@@ -0,0 +1,7 @@
+public static class BoolExtension
+{
+    public static int ToInt(this bool _bool)
+    {
+        return _bool == true ? 1 : 0;
+    }
+}
\ No newline at end of file
diff --git a/Assets/Scripts/Utils/BoolToInteger.cs.meta b/Assets/Scripts/Utils/BoolToInteger.cs.meta
new file mode 100644
index 0000000..3688775
--- /dev/null
+++ b/Assets/Scripts/Utils/BoolToInteger.cs.meta
@@ -0,0 +1,11 @@
+fileFormatVersion: 2
+guid: f48fff3c2eda14d4fba923fe8875f651
+MonoImporter:
+  externalObjects: {}
+  serializedVersion: 2
+  defaultReferences: []
+  executionOrder: 0
+  icon: {instanceID: 0}
+  userData: 
+  assetBundleName: 
+  assetBundleVariant: 
-- 
2.49.0


From c8af0e528415cb8b3b198edc7eb513fdf9133710 Mon Sep 17 00:00:00 2001
From: Krazerleo <krazerleo@gmail.com>
Date: Wed, 4 May 2022 23:50:07 +0700
Subject: [PATCH 02/17] to new git

---
 Assets/Scripts/Bots/CharacterFactory.cs       |  44 +++--
 Assets/Scripts/Bots/TeamEnum.cs               |  11 ++
 Assets/Scripts/Character/Character.cs         |   6 -
 .../Scripts/Character/CharacterCondition.cs   |  32 +++-
 .../Interfaces.meta}                          |   2 +-
 .../Character/Interfaces/ICharacter.cs        |   5 +
 .../Interfaces/ICharacter.cs.meta}            |   2 +-
 .../Character/Interfaces/INpcBaseState.cs     |  17 ++
 .../Interfaces/INpcBaseState.cs.meta}         |   2 +-
 .../Scripts/Character/MovementController.cs   |  70 +++++--
 Assets/Scripts/Character/NPC.cs               | 178 ++++++++++++------
 Assets/Scripts/Character/NPC_State.cs         |  46 -----
 Assets/Scripts/Character/NpcState.cs          |  68 +++++++
 .../{NPC_State.cs.meta => NpcState.cs.meta}   |   0
 Assets/Scripts/Character/Player.cs            |  12 +-
 .../Character/scr_CharacterController.cs      |  64 +++----
 Assets/Scripts/Character/scr_Models.cs        | 120 ++++++------
 Assets/Scripts/Managers/GameManager.cs        | 107 ++++++++---
 Assets/Scripts/Managers/MapManager.cs         |  63 ++++++-
 Assets/Scripts/Managers/TimeManager.cs        |   8 +-
 Assets/Scripts/Misc/FlagZone.cs               |  14 +-
 Assets/Scripts/Misc/NavPoint.cs               |  20 +-
 Assets/Scripts/Misc/Settings.cs               |   4 +-
 Assets/Scripts/Misc/SettingsReader.cs         |  16 +-
 Assets/Scripts/Misc/Statistics.cs             |   9 -
 Assets/Scripts/Pickups/AmmoPickUp.cs          |   8 +-
 Assets/Scripts/Pickups/ArmourPickUp.cs        |   8 +-
 Assets/Scripts/Pickups/HealthPickUp.cs        |   8 +-
 Assets/Scripts/Pickups/IPickable.cs           |   5 +-
 Assets/Scripts/Pickups/PickUpSpawner.cs       |  10 +-
 Assets/Scripts/Sensors/SensorType.cs          |   6 -
 Assets/Scripts/Sensors/Sensors.cs             |   4 -
 Assets/Scripts/Statistics.meta                |   8 +
 Assets/Scripts/Statistics/Logger.cs           |  19 ++
 Assets/Scripts/Statistics/Logger.cs.meta      |  11 ++
 Assets/Scripts/Statistics/StatisticManager.cs |  51 +++++
 .../StatisticManager.cs.meta}                 |   2 +-
 .../Scripts/Utils/SerializableDictionary.cs   |   4 +-
 .../Scripts/Weapons/scr_WeaponController.cs   |  15 +-
 39 files changed, 720 insertions(+), 359 deletions(-)
 rename Assets/Scripts/{Sensors.meta => Character/Interfaces.meta} (77%)
 mode change 100755 => 100644
 create mode 100644 Assets/Scripts/Character/Interfaces/ICharacter.cs
 rename Assets/Scripts/{Sensors/SensorType.cs.meta => Character/Interfaces/ICharacter.cs.meta} (83%)
 mode change 100755 => 100644
 create mode 100644 Assets/Scripts/Character/Interfaces/INpcBaseState.cs
 rename Assets/Scripts/{Sensors/Sensors.cs.meta => Character/Interfaces/INpcBaseState.cs.meta} (83%)
 delete mode 100644 Assets/Scripts/Character/NPC_State.cs
 create mode 100644 Assets/Scripts/Character/NpcState.cs
 rename Assets/Scripts/Character/{NPC_State.cs.meta => NpcState.cs.meta} (100%)
 delete mode 100755 Assets/Scripts/Misc/Statistics.cs
 delete mode 100755 Assets/Scripts/Sensors/SensorType.cs
 delete mode 100644 Assets/Scripts/Sensors/Sensors.cs
 create mode 100644 Assets/Scripts/Statistics.meta
 create mode 100644 Assets/Scripts/Statistics/Logger.cs
 create mode 100644 Assets/Scripts/Statistics/Logger.cs.meta
 create mode 100644 Assets/Scripts/Statistics/StatisticManager.cs
 rename Assets/Scripts/{Misc/Statistics.cs.meta => Statistics/StatisticManager.cs.meta} (91%)
 mode change 100755 => 100644

diff --git a/Assets/Scripts/Bots/CharacterFactory.cs b/Assets/Scripts/Bots/CharacterFactory.cs
index ab18289..0ca7c1d 100644
--- a/Assets/Scripts/Bots/CharacterFactory.cs
+++ b/Assets/Scripts/Bots/CharacterFactory.cs
@@ -1,26 +1,28 @@
 using System.Collections.Generic;
 using UnityEngine;
-using Unity;
 
 public class CharacterFactory : MonoBehaviour
 {
-    private CharacterFactory instance;
-    public CharacterFactory Instance { get { return instance; } }
+    private static CharacterFactory instance;
+    public static CharacterFactory Instance => instance;
 
     [SerializeField] private List<NavPoint> spawnPointsForDefendersTeam;
     [SerializeField] private List<NavPoint> spawnPointsForAttackersTeam;
     [SerializeField] private GameObject AIPrefab;
     [SerializeField] private GameObject PlayerPrefab;
 
-    private List<GameObject> Bots = new List<GameObject>();
-    private GameObject Player;
+    private List<GameObject> bots = new List<GameObject>();
+    public GameObject player { get; private set; }
 
     private void Awake()
     {
         if (instance == null)
             instance = this;
         else
+        {
             Destroy(gameObject);
+            Debug.LogError("Only 1 Instance");
+        }
     }
 
     private void Start()
@@ -53,7 +55,7 @@ public class CharacterFactory : MonoBehaviour
     {
         var gameobject = GameObject.Instantiate(
             typeAi == TypeAI.HumanAI ? PlayerPrefab : AIPrefab,
-            spawnPoint.position,
+            spawnPoint.Position,
             Quaternion.identity);
         gameobject.SetActive(true);
         if (team == Team.Attackers)
@@ -64,35 +66,49 @@ public class CharacterFactory : MonoBehaviour
         if (typeAi == TypeAI.HumanAI)
         {
             gameobject.GetComponent<Player>().GetCharacter.Team = team;
-            Player = gameobject;
+            player = gameobject;
         }
         else
         {
             gameobject.GetComponent<NPC>().GetCharacter.Team = team;
-            gameobject.GetComponent<MovementController>().CurrentNavPoint = spawnPoint;
-            Bots.Add(gameobject);
+            gameobject.GetComponent<MovementController>().PointStartID = spawnPoint.PointId;
+            bots.Add(gameobject);
         }
     }
 
+    public void ReSpawn(ICharacter character, ref Vector3 pos, ref int startPointId)
+    {
+        character.ResetCharacter();
+        var team = character.GetCharacter.Team;
+        NavPoint navPoint;
+        if (team == Team.Attackers)
+            navPoint = spawnPointsForAttackersTeam[Random.Range(0, spawnPointsForAttackersTeam.Count)];
+        else
+            navPoint = spawnPointsForDefendersTeam[Random.Range(0, spawnPointsForDefendersTeam.Count)];
+
+        pos = navPoint.Position;
+        startPointId = navPoint.PointId;
+    }
+
     private void ResetCharacters()
     {
-        foreach (var bot in Bots)
+        foreach (var bot in bots)
         {
             var npc = bot.GetComponent<NPC>();
             npc.ResetCharacter();
             if (npc.GetCharacter.Team == Team.Attackers)
-                bot.transform.position = spawnPointsForAttackersTeam[Random.Range(0, spawnPointsForAttackersTeam.Count)].position;
+                bot.transform.position = spawnPointsForAttackersTeam[Random.Range(0, spawnPointsForAttackersTeam.Count)].Position;
             else
-                bot.transform.position = spawnPointsForDefendersTeam[Random.Range(0, spawnPointsForDefendersTeam.Count)].position;
+                bot.transform.position = spawnPointsForDefendersTeam[Random.Range(0, spawnPointsForDefendersTeam.Count)].Position;
         }
         Player player;
         if (TryGetComponent<Player>(out player))
         {
             player.ResetCharacter();
             if (player.GetCharacter.Team == Team.Attackers)
-                Player.transform.position = spawnPointsForAttackersTeam[Random.Range(0, spawnPointsForAttackersTeam.Count)].position;
+                this.player.transform.position = spawnPointsForAttackersTeam[Random.Range(0, spawnPointsForAttackersTeam.Count)].Position;
             else
-                Player.transform.position = spawnPointsForDefendersTeam[Random.Range(0, spawnPointsForDefendersTeam.Count)].position;
+                this.player.transform.position = spawnPointsForDefendersTeam[Random.Range(0, spawnPointsForDefendersTeam.Count)].Position;
         }
     }
 }
\ No newline at end of file
diff --git a/Assets/Scripts/Bots/TeamEnum.cs b/Assets/Scripts/Bots/TeamEnum.cs
index 68279f5..a2c8a95 100755
--- a/Assets/Scripts/Bots/TeamEnum.cs
+++ b/Assets/Scripts/Bots/TeamEnum.cs
@@ -2,4 +2,15 @@
 {
     Defenders,
     Attackers,
+}
+
+public static class TeamExtension
+{
+    public static Team GetOppositeTeam(this Team team)
+    {
+        if (team == Team.Attackers)
+            return Team.Defenders;
+        else
+            return Team.Attackers;
+    }
 }
\ No newline at end of file
diff --git a/Assets/Scripts/Character/Character.cs b/Assets/Scripts/Character/Character.cs
index 50e7313..3713998 100644
--- a/Assets/Scripts/Character/Character.cs
+++ b/Assets/Scripts/Character/Character.cs
@@ -7,12 +7,6 @@ public class Character
 
     public Character()
     {
-        Debug.Log("init");
         Condition = new CharacterCondition();
     }
-}
-
-public interface ICharacter
-{
-    Character GetCharacter { get; }
 }
\ No newline at end of file
diff --git a/Assets/Scripts/Character/CharacterCondition.cs b/Assets/Scripts/Character/CharacterCondition.cs
index b674222..de60740 100755
--- a/Assets/Scripts/Character/CharacterCondition.cs
+++ b/Assets/Scripts/Character/CharacterCondition.cs
@@ -9,17 +9,17 @@ public class CharacterCondition
     public event Action<int> OnChangeAmmunitionEvent;
 
     private int health;
-    public int HealthPoints 
-    { 
-        get 
-        { 
-            return health; 
-        } 
+    public int HealthPoints
+    {
+        get
+        {
+            return health;
+        }
         private set
         {
             health = value;
             OnChangeHealthEvent?.Invoke(value);
-        } 
+        }
     }
 
     public int GetHealthPointsInQuantile()
@@ -30,7 +30,7 @@ public class CharacterCondition
             return 1;
         else if (health < 75)
             return 2;
-        else return 3;     
+        else return 3;
     }
     private int armour;
     public int ArmourPoints
@@ -45,6 +45,17 @@ public class CharacterCondition
             OnChangeArmourEvent?.Invoke(value);
         }
     }
+    public int GetArmourPointsInQuantile()
+    {
+        if (armour < 25)
+            return 0;
+        else if (armour < 50)
+            return 1;
+        else if (armour < 75)
+            return 2;
+        else return 3;
+    }
+
     private int ammo;
     public int Ammunition
     {
@@ -60,6 +71,11 @@ public class CharacterCondition
     }
 
     public CharacterCondition()
+    {
+        this.Reset();
+    }
+
+    public void Reset()
     {
         var settings = SettingsReader.Instance.GetSettings;
         ammo = settings.MaxAmmo;
diff --git a/Assets/Scripts/Sensors.meta b/Assets/Scripts/Character/Interfaces.meta
old mode 100755
new mode 100644
similarity index 77%
rename from Assets/Scripts/Sensors.meta
rename to Assets/Scripts/Character/Interfaces.meta
index a808c2b..cb4a3aa
--- a/Assets/Scripts/Sensors.meta
+++ b/Assets/Scripts/Character/Interfaces.meta
@@ -1,5 +1,5 @@
 fileFormatVersion: 2
-guid: 5e73ba257bc6b684c86edf9ecfd475ef
+guid: f23b6db3be1e4cd469fd18dfe3e39764
 folderAsset: yes
 DefaultImporter:
   externalObjects: {}
diff --git a/Assets/Scripts/Character/Interfaces/ICharacter.cs b/Assets/Scripts/Character/Interfaces/ICharacter.cs
new file mode 100644
index 0000000..aef14a7
--- /dev/null
+++ b/Assets/Scripts/Character/Interfaces/ICharacter.cs
@@ -0,0 +1,5 @@
+public interface ICharacter
+{
+    Character GetCharacter { get; }
+    void ResetCharacter();
+}
\ No newline at end of file
diff --git a/Assets/Scripts/Sensors/SensorType.cs.meta b/Assets/Scripts/Character/Interfaces/ICharacter.cs.meta
old mode 100755
new mode 100644
similarity index 83%
rename from Assets/Scripts/Sensors/SensorType.cs.meta
rename to Assets/Scripts/Character/Interfaces/ICharacter.cs.meta
index 26f4b58..b53c34e
--- a/Assets/Scripts/Sensors/SensorType.cs.meta
+++ b/Assets/Scripts/Character/Interfaces/ICharacter.cs.meta
@@ -1,5 +1,5 @@
 fileFormatVersion: 2
-guid: 8f76201fe6436164789d10350a0fd6e2
+guid: b6dfb78244ae35c4db1326d5f5b73375
 MonoImporter:
   externalObjects: {}
   serializedVersion: 2
diff --git a/Assets/Scripts/Character/Interfaces/INpcBaseState.cs b/Assets/Scripts/Character/Interfaces/INpcBaseState.cs
new file mode 100644
index 0000000..bf17ad2
--- /dev/null
+++ b/Assets/Scripts/Character/Interfaces/INpcBaseState.cs
@@ -0,0 +1,17 @@
+using UnityEngine;
+
+public interface INpcBaseState
+{
+    NpcEnumState State { get; }
+    bool InCover { get; }
+    bool IsRunning { get; }
+    bool InDirectPoint { get; }
+    float HitChance { get; }
+    float DoDamageChance { get; }
+}
+
+public interface INpcBaseBodyState
+{
+    NpcBodyState State { get; }
+    Vector3 GetPointToHit(GameObject go);
+}
\ No newline at end of file
diff --git a/Assets/Scripts/Sensors/Sensors.cs.meta b/Assets/Scripts/Character/Interfaces/INpcBaseState.cs.meta
similarity index 83%
rename from Assets/Scripts/Sensors/Sensors.cs.meta
rename to Assets/Scripts/Character/Interfaces/INpcBaseState.cs.meta
index 1109bef..f0585e4 100644
--- a/Assets/Scripts/Sensors/Sensors.cs.meta
+++ b/Assets/Scripts/Character/Interfaces/INpcBaseState.cs.meta
@@ -1,5 +1,5 @@
 fileFormatVersion: 2
-guid: 4599c57bc5b1c3945847dead0f9f0ba4
+guid: 58b7e1962495ada4c8e6ee6219c99a20
 MonoImporter:
   externalObjects: {}
   serializedVersion: 2
diff --git a/Assets/Scripts/Character/MovementController.cs b/Assets/Scripts/Character/MovementController.cs
index 334b7d2..2319c08 100644
--- a/Assets/Scripts/Character/MovementController.cs
+++ b/Assets/Scripts/Character/MovementController.cs
@@ -1,24 +1,32 @@
-using System.Linq;
-using System.Collections.Generic;
+using System.Collections.Generic;
+using System.Linq;
 using UnityEngine;
 using UnityEngine.AI;
-using System.Threading.Tasks;
 
 [RequireComponent(typeof(NavMeshAgent))]
 public class MovementController : MonoBehaviour
 {
-    public NavPoint CurrentNavPoint { get; set; }
+    public int PointStartID { get; set; }
+    public int PointEndID { get; private set; }
     public float FlagDistance { get; private set; }
-    private GameObject flag;
-    private const float updateFlagPositionDelay = 5; 
-    [SerializeField] private NavMeshAgent navMeshAgent;
+    private const float updateFlagPositionDelay = 5;
+    private const float updateReachedDestinationDelay = 5;
 
-    private void Start()
+    [SerializeField] private NavMeshAgent navMeshAgent;
+    [SerializeField] private GameObject flag;
+    public float DistanceToGo { get; private set; }
+    public float RemainingDistance => navMeshAgent.remainingDistance;
+    private Dictionary<int, NavPoint> idNavPointDict;
+
+
+    private void Awake()
     {
         navMeshAgent.speed = SettingsReader.Instance.GetSettings.MovementSpeed;
+        idNavPointDict = MapManager.IDToNavPoint;
         InvokeRepeating(nameof(UpdateFlagPosition), 0, updateFlagPositionDelay);
+        InvokeRepeating(nameof(ReachedDestination), 0, updateReachedDestinationDelay);
     }
-
+   
     private void OnDestroy()
     {
         CancelInvoke(nameof(UpdateFlagPosition));
@@ -30,18 +38,46 @@ public class MovementController : MonoBehaviour
     }
 
     public void MoveToRandomPoint()
-    {   
-        Debug.Log(MapManager.navPoints == null);
-        goToNextNavPoint(MapManager.navPoints[Random.Range(0, MapManager.navPoints.Count)]);
+    {
+        Debug.Log(MapManager.NavPoints == null);
+        GoToNextNavPoint(MapManager.NavPoints[Random.Range(0, MapManager.NavPoints.Count)]);
     }
 
-    public List<NavPoint> getPointsCandidate()
+    public List<NavPoint> GetPointsCandidate()
     {
-        return MapManager.navPoints
-            .Where(point => (CurrentNavPoint.position - point.position).magnitude < SettingsReader.Instance.GetSettings.MovementSpeed)
+        return MapManager.NavPoints
+            .Where(point => 
+            (idNavPointDict[PointStartID].Position - point.Position).magnitude < SettingsReader.Instance.GetSettings.MovementDistance)
             .ToList();
     }
 
-    public void goToNextNavPoint(NavPoint destination) =>
-        navMeshAgent.SetDestination(destination.position);
+    public void GoToNextNavPoint(NavPoint destination)
+    {
+        if (navMeshAgent.isStopped == true) navMeshAgent.isStopped = false;
+        PointStartID = PointEndID;
+        PointEndID = destination.PointId;
+        navMeshAgent.SetDestination(destination.Position);
+        DistanceToGo = navMeshAgent.remainingDistance;      
+    }
+
+    public void ReturnToStartPoint()
+    {
+        if (navMeshAgent.isStopped == true) navMeshAgent.isStopped = false;
+        navMeshAgent.SetDestination(idNavPointDict[PointStartID].Position);
+        PointEndID = PointStartID;
+        PointStartID = -1;
+    }
+
+    public void StopOnPath()
+    {
+        navMeshAgent.isStopped = true;
+        PointStartID = -1;
+        PointEndID = -1;
+    }
+
+    public void ReachedDestination()
+    {
+        if ((navMeshAgent.isStopped == false) && (navMeshAgent.velocity.magnitude < 0.1))
+            PointStartID = PointEndID;
+    }
 }
diff --git a/Assets/Scripts/Character/NPC.cs b/Assets/Scripts/Character/NPC.cs
index c738d0b..d47e377 100644
--- a/Assets/Scripts/Character/NPC.cs
+++ b/Assets/Scripts/Character/NPC.cs
@@ -1,108 +1,170 @@
 using System;
-using UnityEngine;
+using System.Collections.Generic;
 using Unity.MLAgents;
-using Unity.MLAgents.Sensors;
 using Unity.MLAgents.Actuators;
+using Unity.MLAgents.Sensors;
+using UnityEngine;
 
-[RequireComponent(typeof(MovementController))]
+[RequireComponent(typeof(MovementController),typeof(BufferSensor))]
 public class NPC : Agent, ICharacter
 {
     [HideInInspector]
-    public Character AgentCharacter;
+    private Character AgentCharacter;
     public CharacterCondition Condition;
-    private FlagZone flagZone;
+    private FlagZone flagZone = null;
 
-    public NPC_BaseState NPC_State { get; private set; }
+    public INpcBaseState NpcState { get; private set; }
+    public INpcBaseBodyState NpcBodyState { get; private set; }
 
     public Character GetCharacter => AgentCharacter;
 
-    private NPC_DirectPointState DirectState;
-    private NPC_InCoverState CoverState;
-    private NPC_RunningState RunningState;
+    private NpcDirectPointState DirectState;
+    private NpcInCoverState CoverState;
+    private NpcRunningState RunningState;
+
+    private NpcStandingState StandingState;
+    private NpcCrouchingState CrouchingState;
 
     private MovementController moveController;
     private BufferSensorComponent bufferSensor;
 
+    private Dictionary<int, NavPoint> navPointIdDict;
+
+    #region UnityEvents and ML
     private void Awake()
     {
-        DirectState = new NPC_DirectPointState();
-        CoverState = new NPC_InCoverState();
-        RunningState = new NPC_RunningState();
-        NPC_State = DirectState;
+        DirectState = new NpcDirectPointState();
+        CoverState = new NpcInCoverState();
+        RunningState = new NpcRunningState();
+        NpcState = DirectState;
+
+        CrouchingState = new NpcCrouchingState();
+        StandingState = new NpcStandingState();
+        NpcBodyState = StandingState;
 
         AgentCharacter = new Character();
         Condition = AgentCharacter.Condition;
 
         moveController = gameObject.GetComponent<MovementController>();
         bufferSensor = gameObject.GetComponent<BufferSensorComponent>();
-    }
-    
 
-    public void ResetCharacter()
+        flagZone = GameObject.FindObjectOfType<FlagZone>();
+        if (flagZone == null)
+            Debug.LogError("Flag Is Not Setted");
+
+        navPointIdDict = MapManager.IDToNavPoint;
+        if (navPointIdDict is null)
+            Debug.LogError("Cant Find Nav Point Dictionary");
+    }
+
+    private void OnDestroy()
     {
-        Condition = new CharacterCondition();
-        EndEpisode();
+        Debug.LogWarning("Pooled object was destroyed");
     }
 
     public override void OnEpisodeBegin()
-    {     
-        NPC_State = DirectState;
+    {
+        NpcState = DirectState;
         flagZone = GameObject.FindObjectOfType<FlagZone>();
     }
 
     public override void CollectObservations(VectorSensor sensor)
     {
-        var candidates = moveController.getPointsCandidate();
+        var candidates = moveController.GetPointsCandidate();
 
-        sensor.AddObservation(Condition.HealthPoints);
-        sensor.AddObservation(Condition.ArmourPoints);
-        sensor.AddObservation(Condition.Ammunition);
-        sensor.AddObservation((int)NPC_State.State);
-        sensor.AddObservation((!flagZone.isNotOccup).ToInt());
+        //common sensors
+        sensor.AddObservation(GameManager.IsHaveSeenByEnemy(AgentCharacter.Team.GetOppositeTeam(),
+                              NpcBodyState.GetPointToHit(gameObject)).ToInt());
         sensor.AddObservation(AgentCharacter.LastTimeHit);
+        sensor.AddObservation((!flagZone.IsNotOccup).ToInt());
         sensor.AddObservation(Condition.GetHealthPointsInQuantile());
+        sensor.AddObservation(Condition.GetArmourPointsInQuantile());
         sensor.AddObservation(candidates.Count);
-        sensor.AddObservation(GameManager.IsEnemyNearby(gameObject.transform.position, AgentCharacter.Team));
-        
+        sensor.AddObservation(moveController.PointStartID);
+        sensor.AddObservation(moveController.PointEndID);
+        //state sensors
+        sensor.AddObservation((int)NpcState.State);
+        sensor.AddObservation((int)NpcBodyState.State);
+        sensor.AddObservation(GameManager.IsEnemyNearby(gameObject.transform.position, AgentCharacter.Team));     
+        sensor.AddObservation(navPointIdDict[moveController.PointStartID].DeathAttr);
+        sensor.AddObservation(navPointIdDict[moveController.PointEndID].DeathAttr);
+        sensor.AddObservation(moveController.FlagDistance);
+
+        //point sensors
         foreach (var point in candidates)
         {
-            Debug.Log((float)moveController.CurrentNavPoint.PointId);
-
             bufferSensor.AppendObservation(new float[] {
-                //1 position in navpointId
-                (float)moveController.CurrentNavPoint.PointId,
-                //2 distance to flag
-                moveController.FlagDistance,
-                //3 death count in point
-                moveController.CurrentNavPoint.DeathAttr,
+                point.DeathAttr,
+                (int)point.navType,
                 //4 flagEnemyDistance
                 GameManager.IsCloserToFlagFromNextNavPoint(point, transform.position).ToInt(),
                 //5 EnemyVsNavPointDistance
-                GameManager.IsCloserToEnemyThanToNextNavPoint(point,transform.position, AgentCharacter.Team).ToInt()
-            }); 
-
-        }
-    }
-
-    public override void Heuristic(in ActionBuffers actionsOut)
-    {
-        var discreteActionsOut = actionsOut.DiscreteActions;
-        if (Input.GetKeyDown(KeyCode.W))
-        {
-            discreteActionsOut[0] = 1;
+                GameManager.IsCloserToEnemyThanToNextNavPoint(point,transform.position, AgentCharacter.Team.GetOppositeTeam()).ToInt(),
+                //6 Have been seen by enemy in this point               
+                GameManager.IsHaveSeenByEnemy(AgentCharacter.Team.GetOppositeTeam(),
+                              point.Position).ToInt()
+        });
         }
     }
 
     public override void OnActionReceived(ActionBuffers actions)
     {
-        if (actions.DiscreteActions[0] == 1)
+        var result = actions.DiscreteActions;
+        if (result[0] == 0)
         {
-            moveController.MoveToRandomPoint();
-            NPC_State = RunningState;
+            if (navPointIdDict[moveController.PointStartID].navType != NavPointType.Cover)
+                return;
+            NpcState = CoverState;
+
+            switch (result[1])
+            {
+                case 0: Peek(); break;
+                case 1: Cover(); break;
+                case 3: Peek(); moveController.GoToNextNavPoint(navPointIdDict[result[2]]); break;
+                case 4: NpcState = DirectState; break;
+                default: throw new ArgumentException("Undefined Action recieved");
+            }
+        }
+        if (result[0] == 1)
+        {
+            if (navPointIdDict[moveController.PointStartID].navType != NavPointType.Direction)
+                return;
+            switch (result[1])
+            {
+                case 0: moveController.GoToNextNavPoint(navPointIdDict[result[2]]);
+                    NpcState = RunningState; break;
+                case 1: NpcState = DirectState; break;
+                default: throw new ArgumentException("Undefined Action recieved");
+            }
+        }
+        if (result[0] == 2)
+        {
+            if (moveController.PointStartID == moveController.PointEndID && moveController.PointEndID != -1)
+                return;            
+            switch (result[1])
+            {
+                case 0: moveController.StopOnPath(); NpcState = DirectState; break;
+                case 1: moveController.ReturnToStartPoint(); NpcState = RunningState; break;
+                default: throw new ArgumentException("Undefined Action recieved");
+            }
         }
     }
+    #endregion
 
-    public event Action<object> OnKilledEvent;
+    public event Action<NpcBodyState> OnChangePosition;
+    private void Peek()
+    {
+        OnChangePosition?.Invoke(global::NpcBodyState.Standing);
+        NpcBodyState = StandingState;
+    }
+
+    private void Cover()
+    {
+        OnChangePosition?.Invoke(global::NpcBodyState.Crouching);
+        NpcBodyState = CrouchingState;
+    }
+
+    public event Action<int, Team> OnDamageRecieved;
     public void GetDamage(float damage)
     {
         AgentCharacter.LastTimeHit = TimeManager.Instance.CurrentTime;
@@ -111,13 +173,17 @@ public class NPC : Agent, ICharacter
 
         if (Condition.HealthPoints < 0)
         {
-            OnKilledEvent?.Invoke(this);
-            moveController.CurrentNavPoint.DeathAttr += 1;
+            MapManager.AddDeathAttributeToPoints(moveController.PointStartID, moveController.PointEndID,
+                                                 moveController.DistanceToGo, moveController.RemainingDistance);
+            var pos = gameObject.transform.position;
+            var id = moveController.PointStartID;
+            CharacterFactory.Instance.ReSpawn(this, ref pos, ref id);
         }
     }
 
-    private void OnDestroy()
+    public void ResetCharacter()
     {
-        Debug.LogWarning("Pooled object was destroyed");
+        Condition.Reset();
+        EndEpisode();
     }
 }
diff --git a/Assets/Scripts/Character/NPC_State.cs b/Assets/Scripts/Character/NPC_State.cs
deleted file mode 100644
index cc2802c..0000000
--- a/Assets/Scripts/Character/NPC_State.cs
+++ /dev/null
@@ -1,46 +0,0 @@
-public enum NPC_EnumState
-{
-    InCover,
-    InDirectPoint,
-    InRunning,
-}
-
-public interface NPC_BaseState
-{
-    NPC_EnumState State { get; }
-    bool InCover { get; }
-    bool IsRunning { get; }
-    bool InDirectPoint { get; }
-    float HitChance { get; }
-    float DoDamageChance { get; }
-}
-
-public class NPC_DirectPointState : NPC_BaseState
-{
-    public bool InCover => false;
-    public bool IsRunning => false;
-    public bool InDirectPoint => false;
-    public float HitChance => SettingsReader.Instance.GetSettings.GetHitChanceInDirectPoint;
-    public float DoDamageChance => SettingsReader.Instance.GetSettings.DoDamageChanceInDirectPoint;
-    public NPC_EnumState State => NPC_EnumState.InDirectPoint;
-}
-
-public class NPC_RunningState : NPC_BaseState
-{
-    public bool InCover => false;
-    public bool IsRunning => true;
-    public bool InDirectPoint => false;
-    public float HitChance => SettingsReader.Instance.GetSettings.GetHitChanceInRunning;
-    public float DoDamageChance => SettingsReader.Instance.GetSettings.DoDamageChanceInRunning;
-    public NPC_EnumState State => NPC_EnumState.InRunning;
-}
-
-public class NPC_InCoverState : NPC_BaseState
-{
-    public bool InCover => true;
-    public bool IsRunning => false;
-    public bool InDirectPoint => false;
-    public float HitChance => SettingsReader.Instance.GetSettings.GetHitChanceInCover;
-    public float DoDamageChance => SettingsReader.Instance.GetSettings.DoDamageChanceInCover;
-    public NPC_EnumState State => NPC_EnumState.InCover;
-}
diff --git a/Assets/Scripts/Character/NpcState.cs b/Assets/Scripts/Character/NpcState.cs
new file mode 100644
index 0000000..51feda6
--- /dev/null
+++ b/Assets/Scripts/Character/NpcState.cs
@@ -0,0 +1,68 @@
+using UnityEngine;
+
+public enum NpcEnumState
+{
+    InCover,
+    InDirectPoint,
+    InRunning,
+}
+
+public enum NpcBodyState
+{
+    Crouching,
+    Standing,
+}
+
+public class NpcCrouchingState : INpcBaseBodyState
+{
+    public NpcBodyState State => NpcBodyState.Crouching;
+
+    public Vector3 GetPointToHit(GameObject go)
+    {
+        MeshRenderer meshRenderer;
+        go.TryGetComponent<MeshRenderer>(out meshRenderer);
+        return meshRenderer.bounds.center;
+    }
+}
+
+public class NpcStandingState : INpcBaseBodyState
+{
+    public NpcBodyState State => NpcBodyState.Standing;
+
+    public Vector3 GetPointToHit(GameObject go)
+    {
+        MeshRenderer meshRenderer;
+        go.TryGetComponent<MeshRenderer>(out meshRenderer);
+        return meshRenderer.bounds.center;
+    }
+}
+
+public class NpcDirectPointState : INpcBaseState
+{
+    public bool InCover => false;
+    public bool IsRunning => false;
+    public bool InDirectPoint => false;
+    public float HitChance => SettingsReader.Instance.GetSettings.GetHitChanceInDirectPoint;
+    public float DoDamageChance => SettingsReader.Instance.GetSettings.DoDamageChanceInDirectPoint;
+    public NpcEnumState State => NpcEnumState.InDirectPoint;
+}
+
+public class NpcRunningState : INpcBaseState
+{
+    public bool InCover => false;
+    public bool IsRunning => true;
+    public bool InDirectPoint => false;
+    public float HitChance => SettingsReader.Instance.GetSettings.GetHitChanceInRunning;
+    public float DoDamageChance => SettingsReader.Instance.GetSettings.DoDamageChanceInRunning;
+    public NpcEnumState State => NpcEnumState.InRunning;
+}
+
+public class NpcInCoverState : INpcBaseState
+{
+    public bool InCover => true;
+    public bool IsRunning => false;
+    public bool InDirectPoint => false;
+    public float HitChance => SettingsReader.Instance.GetSettings.GetHitChanceInCover;
+    public float DoDamageChance => SettingsReader.Instance.GetSettings.DoDamageChanceInCover;
+    public NpcEnumState State => NpcEnumState.InCover;
+}
diff --git a/Assets/Scripts/Character/NPC_State.cs.meta b/Assets/Scripts/Character/NpcState.cs.meta
similarity index 100%
rename from Assets/Scripts/Character/NPC_State.cs.meta
rename to Assets/Scripts/Character/NpcState.cs.meta
diff --git a/Assets/Scripts/Character/Player.cs b/Assets/Scripts/Character/Player.cs
index e593f51..593a326 100644
--- a/Assets/Scripts/Character/Player.cs
+++ b/Assets/Scripts/Character/Player.cs
@@ -15,9 +15,9 @@ public class Player : MonoBehaviour, ICharacter
         Condition = PlayerCharacter.Condition;
     }
 
-    public void ResetCharacter()
+    private void OnDestroy()
     {
-        Condition = new CharacterCondition();
+        Debug.LogWarning("Pooled object was destroyed");
     }
 
     public event Action<object> OnKilledEvent;
@@ -27,12 +27,12 @@ public class Player : MonoBehaviour, ICharacter
         Condition.GiveHealth(-Mathf.RoundToInt(damage * (1 - Condition.ArmourPoints * 0.5f)));
         Condition.GiveArmour(-Mathf.RoundToInt(Mathf.Sqrt(damage) * 5));
 
-        if (Condition.HealthPoints < 0)       
-            OnKilledEvent?.Invoke(this);       
+        if (Condition.HealthPoints < 0)
+            OnKilledEvent?.Invoke(this);
     }
 
-    private void OnDestroy()
+    public void ResetCharacter()
     {
-        Debug.LogWarning("Pooled object was destroyed");
+        Condition = new CharacterCondition();
     }
 }
diff --git a/Assets/Scripts/Character/scr_CharacterController.cs b/Assets/Scripts/Character/scr_CharacterController.cs
index 2e54f22..e4f847d 100755
--- a/Assets/Scripts/Character/scr_CharacterController.cs
+++ b/Assets/Scripts/Character/scr_CharacterController.cs
@@ -1,10 +1,6 @@
-using System;
-using System.Collections;
-using System.Collections.Generic;
-using Unity.Barracuda;
-using UnityEngine;
+using UnityEngine;
 
-using static scr_Models; 
+using static scr_Models;
 
 public class scr_CharacterController : MonoBehaviour
 {
@@ -14,7 +10,7 @@ public class scr_CharacterController : MonoBehaviour
     private Vector2 input_Movement;
     [HideInInspector]
     public Vector2 input_View;
-   
+
     private Vector3 newCameraRotation;
     private Vector3 newCharacterRotation;
 
@@ -22,14 +18,14 @@ public class scr_CharacterController : MonoBehaviour
     public Transform cameraHolder;
     public Transform feetTransform;
 
-    [Header("Settings")] 
+    [Header("Settings")]
     public PlayerSettingsModel playerSettings;
 
     public float ViewClampYMin = -70;
     public float ViewClampYMax = 80;
     public LayerMask playerMask;
-    
-    [Header("Gravity")] 
+
+    [Header("Gravity")]
     public float gravityAmount;
     public float gravityMin;
     private float playerGravity;
@@ -37,14 +33,14 @@ public class scr_CharacterController : MonoBehaviour
     public Vector3 jumpingForce;
     private Vector3 jumpingForceVelocity;
 
-    [Header("Stance")] 
+    [Header("Stance")]
     public PlayerStance playerStance;
     public float playerStanceSmoothing;
     public CharacterStance playerStandStance;
     public CharacterStance playerCrouchStance;
     public CharacterStance playerProneStance;
     private float stanceCheckErrorMargin = 0.05f;
-    
+
     private float cameraHeight;
     private float cameraHeightVelocity;
 
@@ -61,13 +57,13 @@ public class scr_CharacterController : MonoBehaviour
         defaultInput.Character.Movement.performed += e => input_Movement = e.ReadValue<Vector2>();
         defaultInput.Character.View.performed += e => input_View = e.ReadValue<Vector2>();
         defaultInput.Character.Jump.performed += e => Jump();
-        
+
         defaultInput.Character.Crouch.performed += e => Crouch();
         defaultInput.Character.Prone.performed += e => Prone();
-        
+
         defaultInput.Character.Sprint.performed += e => ToggleSprint();
         defaultInput.Character.SprintReleased.performed += e => StopSprint();
-      
+
         defaultInput.Enable();
 
         newCameraRotation = cameraHolder.localRotation.eulerAngles;
@@ -80,7 +76,7 @@ public class scr_CharacterController : MonoBehaviour
         {
             currentWeapon.Initialise(this);
         }
-        
+
     }
 
     private void Update()
@@ -95,10 +91,10 @@ public class scr_CharacterController : MonoBehaviour
     {
         newCharacterRotation.y += playerSettings.ViewXSensetivity * (playerSettings.ViewXInverted ? -input_View.x : input_View.x) * Time.deltaTime;
         transform.localRotation = Quaternion.Euler(newCharacterRotation);
-        
+
         newCameraRotation.x += playerSettings.ViewYSensetivity * (playerSettings.ViewYInverted ? input_View.y : -input_View.y) * Time.deltaTime;
         newCameraRotation.x = Mathf.Clamp(newCameraRotation.x, ViewClampYMin, ViewClampYMax);
-        
+
         cameraHolder.localRotation = Quaternion.Euler(newCameraRotation);
     }
 
@@ -108,7 +104,7 @@ public class scr_CharacterController : MonoBehaviour
         {
             isSprinting = false;
         }
-        
+
         var verticalSpeed = playerSettings.WalkingForwardSpeed;
         var horizontalSpeed = playerSettings.WalkingStrafeSpeed;
 
@@ -117,17 +113,17 @@ public class scr_CharacterController : MonoBehaviour
             verticalSpeed = playerSettings.RunningForwardSpeed;
             horizontalSpeed = playerSettings.RunningStrafeSpeed;
         }
-        
+
         // Effectors
         if (!characterController.isGrounded)
         {
             playerSettings.SpeedEffector = playerSettings.FallingSpeedEffector;
         }
-        else if(playerStance == PlayerStance.Crouch)
+        else if (playerStance == PlayerStance.Crouch)
         {
             playerSettings.SpeedEffector = playerSettings.CrouchSpeedEffector;
-        } 
-        else if(playerStance == PlayerStance.Prone)
+        }
+        else if (playerStance == PlayerStance.Prone)
         {
             playerSettings.SpeedEffector = playerSettings.ProneSpeedEffector;
         }
@@ -135,15 +131,15 @@ public class scr_CharacterController : MonoBehaviour
         {
             playerSettings.SpeedEffector = 1;
         }
-        
+
         verticalSpeed *= playerSettings.SpeedEffector;
         horizontalSpeed *= playerSettings.SpeedEffector;
-        
+
         newMovementSpeed = Vector3.SmoothDamp(newMovementSpeed,
        new Vector3(horizontalSpeed * input_Movement.x * Time.deltaTime,
           0, verticalSpeed * input_Movement.y * Time.deltaTime),
             ref newMovementSpeedVelocity, characterController.isGrounded ? playerSettings.MovementSmoothing : playerSettings.FallingSmoothing);
-        
+
         var MovementSpeed = transform.TransformDirection(newMovementSpeed);
 
         if (playerGravity > gravityMin)
@@ -158,7 +154,7 @@ public class scr_CharacterController : MonoBehaviour
 
         MovementSpeed.y += playerGravity;
         MovementSpeed += jumpingForce * Time.deltaTime;
-        
+
         characterController.Move(MovementSpeed);
     }
 
@@ -179,7 +175,7 @@ public class scr_CharacterController : MonoBehaviour
         {
             stanceHeight = playerProneStance.CameraHeight;
         }
-        
+
         cameraHeight = Mathf.SmoothDamp(cameraHolder.localPosition.y, stanceHeight, ref cameraHeightVelocity, playerStanceSmoothing);
 
         cameraHolder.localPosition = new Vector3(cameraHolder.localPosition.x, cameraHeight, cameraHolder.localPosition.z);
@@ -190,7 +186,7 @@ public class scr_CharacterController : MonoBehaviour
         {
             return;
         }
-        
+
         if (playerStance == PlayerStance.Crouch)
         {
             if (StanceCheck(playerStandStance.StanceCollider.height))
@@ -200,7 +196,7 @@ public class scr_CharacterController : MonoBehaviour
             playerStance = PlayerStance.Stand;
             return;
         }
-        
+
         // Jump
         jumpingForce = Vector3.up * playerSettings.JumpingHeight;
         playerGravity = 0;
@@ -233,8 +229,8 @@ public class scr_CharacterController : MonoBehaviour
     {
         var start = new Vector3(feetTransform.position.x, feetTransform.position.y + characterController.radius + stanceCheckErrorMargin, feetTransform.position.z);
         var end = new Vector3(feetTransform.position.x, feetTransform.position.y - characterController.radius - stanceCheckErrorMargin + stanceCheckheight, feetTransform.position.z);
-        
-        
+
+
         return Physics.CheckCapsule(start, end, characterController.radius, playerMask);
     }
 
@@ -247,7 +243,7 @@ public class scr_CharacterController : MonoBehaviour
         }
         isSprinting = !isSprinting;
     }
-    
+
     private void StopSprint()
     {
         if (playerSettings.SprintingHold)
@@ -255,5 +251,5 @@ public class scr_CharacterController : MonoBehaviour
             isSprinting = false;
         }
     }
-    
+
 }
diff --git a/Assets/Scripts/Character/scr_Models.cs b/Assets/Scripts/Character/scr_Models.cs
index 3aca984..c735db5 100755
--- a/Assets/Scripts/Character/scr_Models.cs
+++ b/Assets/Scripts/Character/scr_Models.cs
@@ -3,73 +3,73 @@ using UnityEngine;
 
 public static class scr_Models
 {
-   #region Player
-   
-   public enum PlayerStance
-   {
-      Stand,
-      Crouch,
-      Prone
-   }
-   
-   [Serializable]
-   public class PlayerSettingsModel
-   {
-      [Header("View Settings")]
-      public float ViewXSensetivity;
-      public float ViewYSensetivity;
+    #region Player
 
-      public bool ViewXInverted;
-      public bool ViewYInverted;
+    public enum PlayerStance
+    {
+        Stand,
+        Crouch,
+        Prone
+    }
 
-      [Header("Movement Settings")] 
-      public bool SprintingHold;
-      public float MovementSmoothing;
-      
-      [Header("Movement - Running")] 
-      public float RunningForwardSpeed;
-      public float RunningStrafeSpeed;
-      
-      [Header("Movement - Walking")] 
-      public float WalkingForwardSpeed;
-      public float WalkingBackwardSpeed;
-      public float WalkingStrafeSpeed;
+    [Serializable]
+    public class PlayerSettingsModel
+    {
+        [Header("View Settings")]
+        public float ViewXSensetivity;
+        public float ViewYSensetivity;
 
-      [Header("Jumping")] 
-      public float JumpingHeight;
-      public float JumpingFalloff;
-      public float FallingSmoothing;
+        public bool ViewXInverted;
+        public bool ViewYInverted;
 
-      [Header("Speed Effectors")] 
-      public float SpeedEffector = 1;
-      public float CrouchSpeedEffector;
-      public float ProneSpeedEffector;
-      public float FallingSpeedEffector;
-   }
+        [Header("Movement Settings")]
+        public bool SprintingHold;
+        public float MovementSmoothing;
 
-   [Serializable]
-   public class CharacterStance
-   {
-      public float CameraHeight;
-      public CapsuleCollider StanceCollider;
-   }
+        [Header("Movement - Running")]
+        public float RunningForwardSpeed;
+        public float RunningStrafeSpeed;
 
-   #endregion
+        [Header("Movement - Walking")]
+        public float WalkingForwardSpeed;
+        public float WalkingBackwardSpeed;
+        public float WalkingStrafeSpeed;
 
-   #region - Weapons -
+        [Header("Jumping")]
+        public float JumpingHeight;
+        public float JumpingFalloff;
+        public float FallingSmoothing;
 
-   [Serializable]
-   public class WeaponSettingsModel
-   {
-      [Header("Sway")]
-      public float SwayAmount;
-      public bool SwayYInverted;
-      public bool SwayXInverted;
-      public float SwaySmoothing;
-      public float SwayResetSmoothing;
-      public float SwayClampX;
-      public float SwayClampY;
-   }
+        [Header("Speed Effectors")]
+        public float SpeedEffector = 1;
+        public float CrouchSpeedEffector;
+        public float ProneSpeedEffector;
+        public float FallingSpeedEffector;
+    }
 
-   #endregion
+    [Serializable]
+    public class CharacterStance
+    {
+        public float CameraHeight;
+        public CapsuleCollider StanceCollider;
+    }
+
+    #endregion
+
+    #region - Weapons -
+
+    [Serializable]
+    public class WeaponSettingsModel
+    {
+        [Header("Sway")]
+        public float SwayAmount;
+        public bool SwayYInverted;
+        public bool SwayXInverted;
+        public float SwaySmoothing;
+        public float SwayResetSmoothing;
+        public float SwayClampX;
+        public float SwayClampY;
+    }
+
+    #endregion
 }
diff --git a/Assets/Scripts/Managers/GameManager.cs b/Assets/Scripts/Managers/GameManager.cs
index b91908d..7720a90 100755
--- a/Assets/Scripts/Managers/GameManager.cs
+++ b/Assets/Scripts/Managers/GameManager.cs
@@ -1,76 +1,123 @@
-using Unity.MLAgents;
+using System;
+using Unity.MLAgents;
 using UnityEngine;
-using System;
 
 public class GameManager : MonoBehaviour
 {
     private static GameManager instance;
-    public static GameManager Instance { get { return instance; } }
+    public static GameManager Instance => instance;
 
-    private static SimpleMultiAgentGroup DefendersTeam = new SimpleMultiAgentGroup();
-    private static SimpleMultiAgentGroup AttackersTeam = new SimpleMultiAgentGroup();
+    private static SimpleMultiAgentGroup defendersTeam = new SimpleMultiAgentGroup();
+    private static SimpleMultiAgentGroup attackersTeam = new SimpleMultiAgentGroup();
 
     private void Awake()
     {
-        if (Instance == null)
+        if (instance is null)
             instance = this;
-        else if (Instance == this)
+        else
+        {
             Destroy(gameObject);
+            Debug.LogError("Only 1 Instance");
+        }
     }
 
     private void Start()
     {
         Academy.Instance.OnEnvironmentReset += ResetScene;
 
-        GlobalEventManager.onCaptureFlag += flagCaptured;
-        GlobalEventManager.onTimeLeft += timeOut;
+        GlobalEventManager.onCaptureFlag += FlagCaptured;
+        GlobalEventManager.onTimeLeft += TimeOut;
 
         var agents = GameObject.FindObjectsOfType<Agent>();
         foreach (var item in agents)
         {
             var agent = item as NPC;
             if (agent.GetCharacter.Team == Team.Attackers)
-                AttackersTeam.RegisterAgent(agent);
+                attackersTeam.RegisterAgent(item);
             else
-                DefendersTeam.RegisterAgent(agent);
+                defendersTeam.RegisterAgent(item);
         }
     }
 
-    public static bool IsCloserToEnemyThanToNextNavPoint(NavPoint navPoint, Vector3 currentTransform, Team team)
+    private static SimpleMultiAgentGroup getAgentList(Team team)
     {
-        SimpleMultiAgentGroup agentGroup;
         if (team == Team.Attackers)
-            agentGroup = AttackersTeam;
+            return attackersTeam;
         else
-            agentGroup = DefendersTeam;
+            return defendersTeam;
+    }
 
-        var distToNavPoint = (currentTransform - navPoint.position).magnitude;
+    public static bool IsCloserToEnemyThanToNextNavPoint(NavPoint navPoint, Vector3 currentTransform, Team oppositeTeam)
+    {
+        var agentGroup = getAgentList(oppositeTeam);      
+
+        var distToNavPoint = (currentTransform - navPoint.Position).magnitude;
         foreach (var agent in agentGroup.GetRegisteredAgents())
             if (distToNavPoint > (currentTransform - agent.transform.position).magnitude)
                 return true;
+        if ((SettingsReader.Instance.GetSettings.HasHumanAttacker == true && oppositeTeam == Team.Attackers) ||
+            (SettingsReader.Instance.GetSettings.HasHumanDefender == true && oppositeTeam == Team.Defenders))
+        {           
+            if (distToNavPoint > (currentTransform - CharacterFactory.Instance.player.transform.position).magnitude)
+                return true;
+        }
         return false;
     }
 
-    public static bool IsEnemyNearby(Vector3 currentTransform, Team team)
+    public static bool IsEnemyNearby(Vector3 currentTransform, Team oppositeTeam)
     {
-        SimpleMultiAgentGroup agentGroup;
-        if (team == Team.Attackers)
-            agentGroup = AttackersTeam;
-        else
-            agentGroup = DefendersTeam;
+        var agentGroup = getAgentList(oppositeTeam);
 
         foreach (var agent in agentGroup.GetRegisteredAgents())
             if ((currentTransform - agent.transform.position).magnitude < SettingsReader.Instance.GetSettings.ViewDistance)
                 return true;
+        if ((SettingsReader.Instance.GetSettings.HasHumanAttacker == true && oppositeTeam == Team.Attackers) ||
+            (SettingsReader.Instance.GetSettings.HasHumanDefender == true && oppositeTeam == Team.Defenders))
+        {
+            if ((currentTransform - CharacterFactory.Instance.player.transform.position).magnitude < SettingsReader.Instance.GetSettings.ViewDistance)
+                return true;
+        }
         return false;
     }
 
     public static bool IsCloserToFlagFromNextNavPoint(NavPoint navPoint, Vector3 currentTransform)
         => navPoint.FlagDistance < (currentTransform - GameObject.FindGameObjectWithTag("Flag").transform.position).magnitude;
 
-    private void flagCaptured(Team team)
+    public static bool IsHaveSeenByEnemy(Team oppositeTeam, Vector3 position)
     {
-        switch(team)
+        var agentGroup = getAgentList(oppositeTeam);
+        RaycastHit rayHit = new RaycastHit();
+        foreach (var agent in agentGroup.GetRegisteredAgents() )
+        {
+            var npc = agent as NPC;
+            if (Physics.Raycast(position,
+                                (npc.NpcBodyState.GetPointToHit(npc.gameObject) - position).normalized,
+                                out rayHit,
+                                SettingsReader.Instance.GetSettings.ViewDistance))
+            {
+                if (rayHit.collider.gameObject.GetComponent<ICharacter>() != null)
+                    return true;
+            }
+        }
+        if ((SettingsReader.Instance.GetSettings.HasHumanAttacker == true && oppositeTeam == Team.Attackers) ||
+            (SettingsReader.Instance.GetSettings.HasHumanDefender == true && oppositeTeam == Team.Defenders))
+        {
+            var player = CharacterFactory.Instance.player;
+            if (Physics.Raycast(position,
+                                (player.GetComponent<MeshRenderer>().bounds.center - position).normalized,
+                                out rayHit,
+                                SettingsReader.Instance.GetSettings.ViewDistance))
+            {
+                if (rayHit.collider.gameObject.GetComponent<ICharacter>() != null)
+                    return true;
+            }
+        }
+        return false;
+    }
+
+    private void FlagCaptured(Team team)
+    {
+        switch (team)
         {
             case Team.Attackers:
                 Debug.Log("Attackers Win");
@@ -78,21 +125,19 @@ public class GameManager : MonoBehaviour
             case Team.Defenders:
                 Debug.Log("Defenders Win");
                 break;
-            default:
-                Debug.LogError("Unexpected Team");
-                break;
         }
+        ResetScene();
     }
 
-    private void timeOut()
+    private void TimeOut()
     {
-        Debug.Log("Time is out");
+        ResetScene();
     }
 
     private void OnDestroy()
     {
-        GlobalEventManager.onCaptureFlag -= flagCaptured;
-        GlobalEventManager.onTimeLeft -= timeOut;
+        GlobalEventManager.onCaptureFlag -= FlagCaptured;
+        GlobalEventManager.onTimeLeft -= TimeOut;
     }
 
     public static event Action OnResetScene;
diff --git a/Assets/Scripts/Managers/MapManager.cs b/Assets/Scripts/Managers/MapManager.cs
index 8c64208..b28dcba 100755
--- a/Assets/Scripts/Managers/MapManager.cs
+++ b/Assets/Scripts/Managers/MapManager.cs
@@ -3,17 +3,62 @@ using UnityEngine;
 
 public class MapManager : MonoBehaviour
 {
-    public static List<NavPoint> navPoints { get; private set; }
-    private void Start()
+    private static MapManager instance;
+    public static MapManager Instance => instance;
+    private static List<NavPoint> navPoints = new List<NavPoint>();
+    private static Dictionary<int, NavPoint> iDToNavPoint = new Dictionary<int, NavPoint>();
+    public static List<NavPoint> NavPoints { get => navPoints; private set => navPoints = value; }
+    public static Dictionary<int, NavPoint> IDToNavPoint { get => iDToNavPoint; private set => iDToNavPoint = value; }
+
+    private void Awake()
     {
-        var i = 0;
-        navPoints = new List<NavPoint>();
-        var navPointsGameObj = GameObject.FindGameObjectsWithTag("Point");
-        foreach (var gameobj in navPointsGameObj)
+        if (instance is null)
+            instance = this;
+        else
         {
-            var navpoint = gameobj.GetComponent<NavPoint>();
-            navpoint.PointId = i; i++;
-            navPoints.Add(navpoint);            
+            Destroy(gameObject);
+            Debug.LogError("Only 1 Instance");
         }
     }
+
+    private void Start()
+    {
+        var navPointSet = GameObject.Find("NavPoint Set");
+        var count = navPointSet.transform.childCount;
+        for (int i=0; i < count; i++)        
+            NavPoints.Add(navPointSet.transform.GetChild(i)
+                .gameObject.GetComponent<NavPoint>());
+
+        NavPointSetToID();
+    }
+
+    private void NavPointSetToID()
+    {
+        int i = 0;
+        foreach (var navPoint in NavPoints)
+        {
+            IDToNavPoint.Add(i, navPoint);
+            navPoint.PointId = i;
+            i++;
+        }
+    }
+
+    public static void AddDeathAttributeToPoints(int startPoint, int endPoint, 
+        float allDistance, float remainingDistance)
+    {
+        var startNavPoint = IDToNavPoint[startPoint];
+        var endNavPoint = IDToNavPoint[endPoint];
+        float coef; 
+        try
+        {
+            coef = remainingDistance / allDistance;
+        }
+        catch (System.ArithmeticException)
+        {
+            Debug.LogError("Path Length is zero");
+            return;
+        }
+        startNavPoint.DeathAttr += 1 - coef;
+        endNavPoint.DeathAttr += coef;
+    }
 }
diff --git a/Assets/Scripts/Managers/TimeManager.cs b/Assets/Scripts/Managers/TimeManager.cs
index ea7fec4..4aa32a1 100755
--- a/Assets/Scripts/Managers/TimeManager.cs
+++ b/Assets/Scripts/Managers/TimeManager.cs
@@ -1,6 +1,4 @@
-using System.Collections;
-using System.Collections.Generic;
-using UnityEngine;
+using UnityEngine;
 
 public class TimeManager : MonoBehaviour
 {
@@ -17,12 +15,14 @@ public class TimeManager : MonoBehaviour
         }
         else
         {
-            Debug.LogError("Only one Instance");
+            Debug.LogError("Only 1 Instance");
             Destroy(gameObject);
         }
     }
     void Update()
     {
         CurrentTime += Time.deltaTime;
+        if (CurrentTime > SettingsReader.Instance.GetSettings.TimeOut)
+            GlobalEventManager.SendTimeout();
     }
 }
diff --git a/Assets/Scripts/Misc/FlagZone.cs b/Assets/Scripts/Misc/FlagZone.cs
index c6328dd..c66d5bb 100755
--- a/Assets/Scripts/Misc/FlagZone.cs
+++ b/Assets/Scripts/Misc/FlagZone.cs
@@ -1,6 +1,4 @@
-using System.Collections;
-using System.Collections.Generic;
-using UnityEngine;
+using UnityEngine;
 
 public class FlagZone : MonoBehaviour
 {
@@ -9,8 +7,8 @@ public class FlagZone : MonoBehaviour
     public float TimeStayDefenders { get; private set; }
     private int occupDefenders;
     private int occupAttackers;
-    public bool isOccupBoth => (occupDefenders>0) && (occupAttackers>0);
-    public bool isNotOccup => (occupDefenders == 0) && (occupAttackers == 0);
+    public bool IsOccupBoth => (occupDefenders > 0) && (occupAttackers > 0);
+    public bool IsNotOccup => (occupDefenders == 0) && (occupAttackers == 0);
     private float timeForWin;
 
     private void Start()
@@ -24,7 +22,7 @@ public class FlagZone : MonoBehaviour
     }
     private void OnTriggerEnter(Collider other)
     {
-        switch(other.tag)
+        switch (other.tag)
         {
             case "Defender":
                 occupDefenders++;
@@ -54,7 +52,7 @@ public class FlagZone : MonoBehaviour
     }
     private void Update()
     {
-        if (isOccupBoth || isNotOccup)
+        if (IsOccupBoth || IsNotOccup)
         {
             TimeStayAttackers = 0;
             TimeStayDefenders = 0;
@@ -64,7 +62,7 @@ public class FlagZone : MonoBehaviour
         {
             TimeStayAttackers += Time.deltaTime;
             if (TimeStayAttackers > timeForWin)
-                GlobalEventManager.SendCaptureFlag(Team.Attackers); 
+                GlobalEventManager.SendCaptureFlag(Team.Attackers);
         }
         else
         {
diff --git a/Assets/Scripts/Misc/NavPoint.cs b/Assets/Scripts/Misc/NavPoint.cs
index 2b43e84..a70fd61 100755
--- a/Assets/Scripts/Misc/NavPoint.cs
+++ b/Assets/Scripts/Misc/NavPoint.cs
@@ -1,20 +1,28 @@
-using System.Collections;
-using System.Collections.Generic;
+using System.Collections.Generic;
+using System;
 using UnityEngine;
 
+public enum NavPointType
+{
+    Cover,
+    Direction,
+}
+
+
 public class NavPoint : MonoBehaviour
 {
-    public Vector3 position => gameObject.transform.position;
+    public Vector3 Position => gameObject.transform.position;
     public float FlagDistance { get; private set; }
 
+    public NavPointType navType = NavPointType.Direction;
+
     [HideInInspector]
-    public int? PointId;
+    public int PointId = 0;
     public float DeathAttr = 0;
     public List<Vector3> EnemiesSeen = new List<Vector3>();
-    //Here other attributes;
 
     private void Start()
     {
-        FlagDistance = (GameObject.FindGameObjectWithTag("Flag").transform.position - position).magnitude;               
+        FlagDistance = (GameObject.FindGameObjectWithTag("Flag").transform.position - Position).magnitude;
     }
 }
diff --git a/Assets/Scripts/Misc/Settings.cs b/Assets/Scripts/Misc/Settings.cs
index b1b7ea8..4e333fc 100755
--- a/Assets/Scripts/Misc/Settings.cs
+++ b/Assets/Scripts/Misc/Settings.cs
@@ -1,6 +1,6 @@
 using UnityEngine;
 
-[CreateAssetMenu(fileName ="Game Settings", menuName = "Game/Settings", order = 51)]
+[CreateAssetMenu(fileName = "Game Settings", menuName = "Game/Settings", order = 51)]
 public class Settings : ScriptableObject
 {
     public bool IsTesting;
@@ -36,4 +36,6 @@ public class Settings : ScriptableObject
     public float DoDamageChanceInDirectPoint;
     public float DoDamageChanceInRunning;
     public float DoDamageChanceInCover;
+
+    public float CrouchingCoefficient;
 }
diff --git a/Assets/Scripts/Misc/SettingsReader.cs b/Assets/Scripts/Misc/SettingsReader.cs
index ab60b87..9e709ad 100755
--- a/Assets/Scripts/Misc/SettingsReader.cs
+++ b/Assets/Scripts/Misc/SettingsReader.cs
@@ -1,17 +1,21 @@
-using System.Collections;
-using System.Collections.Generic;
-using UnityEngine;
+using UnityEngine;
 
 public class SettingsReader : MonoBehaviour
 {
     private static SettingsReader instance;
-    public static SettingsReader Instance { get { return instance; } }
+    public static SettingsReader Instance => instance;
 
     private void Awake()
     {
-        instance = this;
+        if (instance is null)
+            instance = this;
+        else
+        {
+            Destroy(gameObject);
+            Debug.LogError("Only 1 Instance");
+        }
     }
 
     [SerializeField] private Settings gameSettings;
-    public Settings GetSettings { get { return gameSettings; } }
+    public Settings GetSettings => gameSettings;
 }
diff --git a/Assets/Scripts/Misc/Statistics.cs b/Assets/Scripts/Misc/Statistics.cs
deleted file mode 100755
index 98be783..0000000
--- a/Assets/Scripts/Misc/Statistics.cs
+++ /dev/null
@@ -1,9 +0,0 @@
-using UnityEngine;
-
-public class Statistics : MonoBehaviour
-{
-    private void Start()
-    {
-        
-    }
-}
diff --git a/Assets/Scripts/Pickups/AmmoPickUp.cs b/Assets/Scripts/Pickups/AmmoPickUp.cs
index 8439c21..e7af1ce 100755
--- a/Assets/Scripts/Pickups/AmmoPickUp.cs
+++ b/Assets/Scripts/Pickups/AmmoPickUp.cs
@@ -1,5 +1,4 @@
-using System;
-using UnityEngine;
+using UnityEngine;
 
 [RequireComponent(typeof(BoxCollider))]
 public class AmmoPickUp : MonoBehaviour, IPickable
@@ -11,6 +10,11 @@ public class AmmoPickUp : MonoBehaviour, IPickable
         PickObject(other.gameObject);
     }
 
+    private void OnDestroy()
+    {
+        Debug.LogWarning("Pooled object was destroyed");
+    }
+
     public void PickObject(GameObject obj)
     {
         obj.GetComponent<ICharacter>()?.GetCharacter.Condition.TakeAmmo(SettingsReader.Instance.GetSettings.AmmunitionPickupAmount);
diff --git a/Assets/Scripts/Pickups/ArmourPickUp.cs b/Assets/Scripts/Pickups/ArmourPickUp.cs
index c68e491..b5303f6 100755
--- a/Assets/Scripts/Pickups/ArmourPickUp.cs
+++ b/Assets/Scripts/Pickups/ArmourPickUp.cs
@@ -1,5 +1,4 @@
-using System;
-using UnityEngine;
+using UnityEngine;
 
 [RequireComponent(typeof(BoxCollider))]
 public class ArmourPickUp : MonoBehaviour, IPickable
@@ -11,6 +10,11 @@ public class ArmourPickUp : MonoBehaviour, IPickable
         PickObject(other.gameObject);
     }
 
+    private void OnDestroy()
+    {
+        Debug.LogWarning("Pooled object was destroyed");
+    }
+
     public void PickObject(GameObject obj)
     {
         obj.GetComponent<ICharacter>()?.GetCharacter.Condition.GiveArmour(SettingsReader.Instance.GetSettings.ArmourPickupAmount);
diff --git a/Assets/Scripts/Pickups/HealthPickUp.cs b/Assets/Scripts/Pickups/HealthPickUp.cs
index 36f81ba..ba8d136 100755
--- a/Assets/Scripts/Pickups/HealthPickUp.cs
+++ b/Assets/Scripts/Pickups/HealthPickUp.cs
@@ -1,5 +1,4 @@
-using System;
-using UnityEngine;
+using UnityEngine;
 
 [RequireComponent(typeof(BoxCollider))]
 public class HealthPickUp : MonoBehaviour, IPickable
@@ -11,6 +10,11 @@ public class HealthPickUp : MonoBehaviour, IPickable
         PickObject(other.gameObject);
     }
 
+    private void OnDestroy()
+    {
+        Debug.LogWarning("Pooled object was destroyed");
+    }
+
     public void PickObject(GameObject obj)
     {
         obj.GetComponent<ICharacter>()?.GetCharacter.Condition.GiveHealth(SettingsReader.Instance.GetSettings.HealthPickupAmount);
diff --git a/Assets/Scripts/Pickups/IPickable.cs b/Assets/Scripts/Pickups/IPickable.cs
index fb218d0..37cced3 100755
--- a/Assets/Scripts/Pickups/IPickable.cs
+++ b/Assets/Scripts/Pickups/IPickable.cs
@@ -1,7 +1,6 @@
-using System;
-using UnityEngine;
+using UnityEngine;
 public interface IPickable
 {
-    PickUpType type { get; } 
+    PickUpType type { get; }
     void PickObject(GameObject obj);
 }
\ No newline at end of file
diff --git a/Assets/Scripts/Pickups/PickUpSpawner.cs b/Assets/Scripts/Pickups/PickUpSpawner.cs
index 7a29705..f9c320b 100755
--- a/Assets/Scripts/Pickups/PickUpSpawner.cs
+++ b/Assets/Scripts/Pickups/PickUpSpawner.cs
@@ -44,25 +44,25 @@ public class PickUpSpawner : MonoBehaviour
 
     private IEnumerator SpawnNewPickUps()
     {
-        while(true)
+        while (true)
         {
             GameObject item;
-            if(IsDisableCheck(out item))
+            if (IsDisableCheck(out item))
             {
                 yield return new WaitForSeconds(3);
                 if (item != null)
                 {
-                    item.transform.position = spawnPoints[Random.Range(0, spawnPoints.Count)].position;
+                    item.transform.position = spawnPoints[Random.Range(0, spawnPoints.Count)].Position;
                     item.SetActive(true);
                 }
             }
-            yield return new  WaitForSeconds(2);
+            yield return new WaitForSeconds(2);
         }
     }
 
     private bool IsDisableCheck(out GameObject gameobj)
     {
-        foreach(var pick in pickups)
+        foreach (var pick in pickups)
         {
             if (!pick.activeInHierarchy)
             {
diff --git a/Assets/Scripts/Sensors/SensorType.cs b/Assets/Scripts/Sensors/SensorType.cs
deleted file mode 100755
index 8d85790..0000000
--- a/Assets/Scripts/Sensors/SensorType.cs
+++ /dev/null
@@ -1,6 +0,0 @@
-public enum SensorType
-{
-    Visual,
-    Sound,
-    Other
-}
diff --git a/Assets/Scripts/Sensors/Sensors.cs b/Assets/Scripts/Sensors/Sensors.cs
deleted file mode 100644
index b38d85f..0000000
--- a/Assets/Scripts/Sensors/Sensors.cs
+++ /dev/null
@@ -1,4 +0,0 @@
-using System.Collections.Generic;
-using Unity.MLAgents.Sensors;
-
-
diff --git a/Assets/Scripts/Statistics.meta b/Assets/Scripts/Statistics.meta
new file mode 100644
index 0000000..4e72120
--- /dev/null
+++ b/Assets/Scripts/Statistics.meta
@@ -0,0 +1,8 @@
+fileFormatVersion: 2
+guid: 3a9f7f0a9faf11f49a433480722bffc5
+folderAsset: yes
+DefaultImporter:
+  externalObjects: {}
+  userData: 
+  assetBundleName: 
+  assetBundleVariant: 
diff --git a/Assets/Scripts/Statistics/Logger.cs b/Assets/Scripts/Statistics/Logger.cs
new file mode 100644
index 0000000..e293d25
--- /dev/null
+++ b/Assets/Scripts/Statistics/Logger.cs
@@ -0,0 +1,19 @@
+using System.IO;
+using UnityEngine;
+
+public class Logger
+{
+    private const string directory = "/Logs/";
+    private const string baseName = "Log#";
+
+    public static void SaveLog<T>(T objToSerialize)
+    {
+        string dir = Application.persistentDataPath + directory;
+        if (!Directory.Exists(dir))
+            Directory.CreateDirectory(dir);
+
+        var logName = baseName + (Directory.GetFiles(dir).Length + 1).ToString();
+        string json = JsonUtility.ToJson(objToSerialize);
+        File.WriteAllText(dir + logName, json);
+    }
+}
\ No newline at end of file
diff --git a/Assets/Scripts/Statistics/Logger.cs.meta b/Assets/Scripts/Statistics/Logger.cs.meta
new file mode 100644
index 0000000..e455173
--- /dev/null
+++ b/Assets/Scripts/Statistics/Logger.cs.meta
@@ -0,0 +1,11 @@
+fileFormatVersion: 2
+guid: b3a1cec894fa98b4bbe20470f1e316c4
+MonoImporter:
+  externalObjects: {}
+  serializedVersion: 2
+  defaultReferences: []
+  executionOrder: 0
+  icon: {instanceID: 0}
+  userData: 
+  assetBundleName: 
+  assetBundleVariant: 
diff --git a/Assets/Scripts/Statistics/StatisticManager.cs b/Assets/Scripts/Statistics/StatisticManager.cs
new file mode 100644
index 0000000..399e7fb
--- /dev/null
+++ b/Assets/Scripts/Statistics/StatisticManager.cs
@@ -0,0 +1,51 @@
+using UnityEngine;
+
+internal class Log
+{
+    public int damageTakenByDefs = 0;
+    public int damageTakenByAtc = 0;
+
+    public int AtcWin = 0;
+    public int DefWin = 0;
+
+    public int TimeOuts = 0;
+}
+
+public class StatisticManager : MonoBehaviour
+{
+    private Log log = new Log();
+    private void Awake()
+    {
+        foreach (var npc in GameObject.FindObjectsOfType<NPC>())
+            npc.OnDamageRecieved += RegisterDamage;
+
+        GlobalEventManager.onCaptureFlag += RegisterWin;
+        GlobalEventManager.onTimeLeft += RegisterTimeOut;
+    }
+
+    private void RegisterDamage(int damage, Team team)
+    {
+        if (team == Team.Attackers)
+            log.damageTakenByAtc += damage;
+        else
+            log.damageTakenByDefs += damage;
+    }
+
+    private void RegisterWin(Team team)
+    {
+        if (team == Team.Attackers)
+            log.AtcWin += 1;
+        else
+            log.DefWin += 1;
+    }
+
+    private void RegisterTimeOut()
+    {
+        log.TimeOuts += 1;
+    }
+
+    private void OnApplicationQuit()
+    {
+        Logger.SaveLog<Log>(log);
+    }
+}
diff --git a/Assets/Scripts/Misc/Statistics.cs.meta b/Assets/Scripts/Statistics/StatisticManager.cs.meta
old mode 100755
new mode 100644
similarity index 91%
rename from Assets/Scripts/Misc/Statistics.cs.meta
rename to Assets/Scripts/Statistics/StatisticManager.cs.meta
index b8be93a..3a27c34
--- a/Assets/Scripts/Misc/Statistics.cs.meta
+++ b/Assets/Scripts/Statistics/StatisticManager.cs.meta
@@ -4,7 +4,7 @@ MonoImporter:
   externalObjects: {}
   serializedVersion: 2
   defaultReferences: []
-  executionOrder: 0
+  executionOrder: 300
   icon: {instanceID: 0}
   userData: 
   assetBundleName: 
diff --git a/Assets/Scripts/Utils/SerializableDictionary.cs b/Assets/Scripts/Utils/SerializableDictionary.cs
index c0877b9..8cb258e 100755
--- a/Assets/Scripts/Utils/SerializableDictionary.cs
+++ b/Assets/Scripts/Utils/SerializableDictionary.cs
@@ -1,10 +1,10 @@
 using System;
-using System.Linq;
 using System.Collections;
 using System.Collections.Generic;
 using System.Diagnostics;
-using UnityEngine;
+using System.Linq;
 using UnityEditor;
+using UnityEngine;
 using UnityObject = UnityEngine.Object;
 
 [Serializable, DebuggerDisplay("Count = {Count}")]
diff --git a/Assets/Scripts/Weapons/scr_WeaponController.cs b/Assets/Scripts/Weapons/scr_WeaponController.cs
index 2fc59e5..613fc61 100755
--- a/Assets/Scripts/Weapons/scr_WeaponController.cs
+++ b/Assets/Scripts/Weapons/scr_WeaponController.cs
@@ -1,17 +1,16 @@
-using System;
-using UnityEngine;
+using UnityEngine;
 using static scr_Models;
 public class scr_WeaponController : MonoBehaviour
 {
     private scr_CharacterController characterController;
-    [Header("Settings")] 
+    [Header("Settings")]
     public WeaponSettingsModel settings;
 
     private bool isInitialised;
 
     Vector3 newWeaponRotation;
     Vector3 newWeaponRotationVelocity;
-    
+
     Vector3 targetWeaponRotation;
     Vector3 targetWeaponRotationVelocity;
 
@@ -32,17 +31,17 @@ public class scr_WeaponController : MonoBehaviour
         {
             return;
         }
-        
+
         targetWeaponRotation.y += settings.SwayAmount * (settings.SwayXInverted ? -characterController.input_View.x : characterController.input_View.x) * Time.deltaTime;
-        targetWeaponRotation.x += settings.SwayAmount  * (settings.SwayYInverted ? characterController.input_View.y : -characterController.input_View.y) * Time.deltaTime;
+        targetWeaponRotation.x += settings.SwayAmount * (settings.SwayYInverted ? characterController.input_View.y : -characterController.input_View.y) * Time.deltaTime;
         //newWeaponRotation.x = Mathf.Clamp(newWeaponRotation.x, ViewClampYMin, ViewClampYMax);
-        
+
         targetWeaponRotation.x = Mathf.Clamp(targetWeaponRotation.x, -settings.SwayClampX, settings.SwayClampX);
         targetWeaponRotation.y = Mathf.Clamp(targetWeaponRotation.y, -settings.SwayClampY, settings.SwayClampY);
 
         targetWeaponRotation = Vector3.SmoothDamp(targetWeaponRotation, Vector3.zero, ref targetWeaponRotationVelocity, settings.SwayResetSmoothing);
         newWeaponRotation = Vector3.SmoothDamp(newWeaponRotation, targetWeaponRotation, ref newWeaponRotationVelocity, settings.SwaySmoothing);
-        
+
         transform.localRotation = Quaternion.Euler(newWeaponRotation);
     }
 }
-- 
2.49.0


From 3420d14cdc0bc595b4e5b0eee3bb8d7b508fb4b7 Mon Sep 17 00:00:00 2001
From: Krazerleo <krazerleo@gmail.com>
Date: Wed, 4 May 2022 23:57:38 +0700
Subject: [PATCH 03/17] merge

---
 .../Timers/Greatest_map_ever_timers.json      |  2 +-
 Assets/Prefabs/Bot.prefab                     |  3 +-
 .../Greatest_map_ever/Greatest_map_ever.unity | 29 ++++++++++-
 Assets/Settings/Game Settings.asset           | 50 ++++++++++---------
 4 files changed, 57 insertions(+), 27 deletions(-)

diff --git a/Assets/ML-Agents/Timers/Greatest_map_ever_timers.json b/Assets/ML-Agents/Timers/Greatest_map_ever_timers.json
index 9a97eb3..b96488e 100644
--- a/Assets/ML-Agents/Timers/Greatest_map_ever_timers.json
+++ b/Assets/ML-Agents/Timers/Greatest_map_ever_timers.json
@@ -1 +1 @@
-{"count":1,"self":16.593504,"total":17.166309,"children":{"InitializeActuators":{"count":2,"self":0.0019996,"total":0.0019996,"children":null},"InitializeSensors":{"count":2,"self":0.0030004,"total":0.0030004,"children":null},"AgentSendState":{"count":612,"self":0.016998,"total":0.549809,"children":{"CollectObservations":{"count":1224,"self":0.0230042,"total":0.0230042,"children":null},"WriteActionMask":{"count":1224,"self":0.0060088,"total":0.0060088,"children":null},"RequestDecision":{"count":1224,"self":0.503798,"total":0.503798,"children":null}}},"DecideAction":{"count":612,"self":0.0089991,"total":0.0089991,"children":null},"AgentAct":{"count":612,"self":0.006997,"total":0.006997,"children":null}},"gauges":{"My Behavior.CumulativeReward":{"count":14,"max":0,"min":0,"runningAverage":0,"value":0,"weightedAverage":0}},"metadata":{"timer_format_version":"0.1.0","start_time_seconds":"1650253950","unity_version":"2019.4.35f1","command_line_arguments":"C:\\Program Files\\unityeditorfolder\\2019.4.35f1\\Editor\\Unity.exe -projectpath F:\\SigmaRiskManagment\\real shooter Git Version -useHub -hubIPC -cloudEnvironment production -licensingIpc LicenseClient-1IWpvtxiu_rvPpHhMWpzt -hubSessionId b1d8b690-be9f-11ec-92bc-6fd1276b6775 -accessToken D1AF5mitRE4Vh3s6p7rpmGuqcqvTsZJKmoGtSNfyNNs009f","communication_protocol_version":"1.5.0","com.unity.ml-agents_version":"2.0.1","scene_name":"Greatest_map_ever","end_time_seconds":"1650253967"}}
\ No newline at end of file
+{"count":1,"self":5.2745576,"total":5.3091238,"children":{"InitializeActuators":{"count":2,"self":0.0019998999999999998,"total":0.0019998999999999998,"children":null},"InitializeSensors":{"count":2,"self":0.0030009999999999998,"total":0.0030009999999999998,"children":null},"AgentSendState":{"count":18,"self":0.0019986,"total":0.0175794,"children":{"CollectObservations":{"count":36,"self":0.0009994,"total":0.0009994,"children":null},"WriteActionMask":{"count":36,"self":0.0005783,"total":0.0005783,"children":null},"RequestDecision":{"count":36,"self":0.0140031,"total":0.0140031,"children":null}}},"DecideAction":{"count":18,"self":0.0089876,"total":0.0089876,"children":null},"AgentAct":{"count":18,"self":0.0009983,"total":0.0009983,"children":null}},"gauges":{"My Behavior.CumulativeReward":{"count":2,"max":0,"min":0,"runningAverage":0,"value":0,"weightedAverage":0}},"metadata":{"timer_format_version":"0.1.0","start_time_seconds":"1651670021","unity_version":"2019.4.35f1","command_line_arguments":"C:\\Program Files\\unityeditorfolder\\2019.4.35f1\\Editor\\Unity.exe -projectpath F:\\SigmaRiskManagment\\real shooter Git Version -useHub -hubIPC -cloudEnvironment production -licensingIpc LicenseClient-9THNgaHTf51SEKo5URf9r -hubSessionId bf0b1ec0-c926-11ec-9066-7725c5249b8e -accessToken pLNs79-R0y77otVjQKfPq_jck-T6ncSnk21wIQRhdZw009f","communication_protocol_version":"1.5.0","com.unity.ml-agents_version":"2.0.1","scene_name":"Greatest_map_ever","end_time_seconds":"1651670026"}}
\ No newline at end of file
diff --git a/Assets/Prefabs/Bot.prefab b/Assets/Prefabs/Bot.prefab
index eedbf35..619d624 100755
--- a/Assets/Prefabs/Bot.prefab
+++ b/Assets/Prefabs/Bot.prefab
@@ -143,7 +143,7 @@ MonoBehaviour:
   m_Name: 
   m_EditorClassIdentifier: 
   m_BrainParameters:
-    VectorObservationSize: 1
+    VectorObservationSize: 9
     NumStackedVectorObservations: 1
     m_ActionSpec:
       m_NumContinuousActions: 0
@@ -195,6 +195,7 @@ MonoBehaviour:
   m_Name: 
   m_EditorClassIdentifier: 
   navMeshAgent: {fileID: 8656710265340117963}
+  flag: {fileID: 6818223691859422291, guid: 1685c1d9ce4ab174f95c646b1826010b, type: 3}
 --- !u!114 &1208561866453126566
 MonoBehaviour:
   m_ObjectHideFlags: 0
diff --git a/Assets/Scenes/Greatest_map_ever/Greatest_map_ever.unity b/Assets/Scenes/Greatest_map_ever/Greatest_map_ever.unity
index 41d83a6..3bcfbee 100755
--- a/Assets/Scenes/Greatest_map_ever/Greatest_map_ever.unity
+++ b/Assets/Scenes/Greatest_map_ever/Greatest_map_ever.unity
@@ -259,6 +259,8 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  navType: 1
+  PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
 --- !u!23 &140697607
@@ -353,6 +355,8 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  navType: 1
+  PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
 --- !u!23 &293522541
@@ -1315,6 +1319,8 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  navType: 1
+  PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
 --- !u!23 &1116745545
@@ -1477,7 +1483,7 @@ Mesh:
   m_CorrespondingSourceObject: {fileID: 0}
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
-  m_Name: pb_Mesh17416
+  m_Name: pb_Mesh16900
   serializedVersion: 10
   m_SubMeshes:
   - serializedVersion: 2
@@ -1763,6 +1769,8 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  navType: 1
+  PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
 --- !u!23 &1345085343
@@ -1857,6 +1865,8 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  navType: 1
+  PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
 --- !u!23 &1488699524
@@ -2077,6 +2087,8 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  navType: 1
+  PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
 --- !u!23 &1663305224
@@ -2141,6 +2153,7 @@ GameObject:
   - component: {fileID: 1858987090}
   - component: {fileID: 1858987088}
   - component: {fileID: 1858987085}
+  - component: {fileID: 1858987091}
   m_Layer: 0
   m_Name: Game
   m_TagString: Untagged
@@ -2253,6 +2266,18 @@ MonoBehaviour:
     type: 3}
   PlayerPrefab: {fileID: 5245491127989480125, guid: 80f6c1c85e5daed4c96c70205ed5503d,
     type: 3}
+--- !u!114 &1858987091
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  m_GameObject: {fileID: 1858987083}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: bf3fe86787bfb0c4b8751fe495148ede, type: 3}
+  m_Name: 
+  m_EditorClassIdentifier: 
 --- !u!1 &1871257865
 GameObject:
   m_ObjectHideFlags: 0
@@ -2506,6 +2531,8 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  navType: 1
+  PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
 --- !u!23 &2004854095
diff --git a/Assets/Settings/Game Settings.asset b/Assets/Settings/Game Settings.asset
index f1a62c0..fc07488 100755
--- a/Assets/Settings/Game Settings.asset	
+++ b/Assets/Settings/Game Settings.asset	
@@ -12,27 +12,29 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: e2c47233b9062c84482336b145c6891b, type: 3}
   m_Name: Game Settings
   m_EditorClassIdentifier: 
-  isTesting: 1
-  timeToWin: 5
-  timeOut: 1600
-  movementDistance: 50
-  movementSpeed: 3
-  defTeamAI: 0
-  atcTeamAI: 0
-  numOfDefenders: 1
-  numOfAttackers: 1
-  hasHumanDefender: 1
-  hasHumanAttacker: 0
-  healthPickupAmount: 50
-  armourPickupAmount: 50
-  ammunitionPickupAmount: 120
-  pickupsAmount: 2
-  maxHealth: 0
-  maxArmour: 0
-  maxAmmo: 0
-  GetHitChanceInDirectPoint: 0
-  GetHitChanceInRunning: 0
-  GetHitChanceInCover: 0
-  DoDamageChanceInDirectPoint: 0
-  DoDamageChanceInRunning: 0
-  DoDamageChanceInCover: 0
+  IsTesting: 0
+  TimeToWin: 15
+  TimeOut: 3600
+  MovementDistance: 50
+  MovementSpeed: 5
+  DefTeamAI: 3
+  AtcTeamAI: 3
+  NumOfDefenders: 1
+  NumOfAttackers: 1
+  HasHumanDefender: 0
+  HasHumanAttacker: 0
+  HealthPickupAmount: 50
+  ArmourPickupAmount: 50
+  AmmunitionPickupAmount: 60
+  PickupsAmount: 0
+  MaxHealth: 100
+  MaxArmour: 100
+  MaxAmmo: 360
+  ViewDistance: 100
+  GetHitChanceInDirectPoint: 50
+  GetHitChanceInRunning: 25
+  GetHitChanceInCover: 20
+  DoDamageChanceInDirectPoint: 70
+  DoDamageChanceInRunning: 30
+  DoDamageChanceInCover: 25
+  CrouchingCoefficient: 1.4
-- 
2.49.0


From e689927b730e87ecd8ba02559bde925a7df5b09d Mon Sep 17 00:00:00 2001
From: Krazerleo <krazerleo@gmail.com>
Date: Thu, 5 May 2022 00:55:51 +0700
Subject: [PATCH 04/17] dadwa

---
 StyleCop.Cache | 7477 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 7477 insertions(+)
 create mode 100644 StyleCop.Cache

diff --git a/StyleCop.Cache b/StyleCop.Cache
new file mode 100644
index 0000000..edcafe4
--- /dev/null
+++ b/StyleCop.Cache
@@ -0,0 +1,7477 @@
+<stylecopresultscache>
+  <version>12</version>
+  <sourcecode name="CharacterFactory.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.24 01:08:14.230</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="UsingDirectivesMustBeOrderedAlphabeticallyByNamespace" ruleCheckId="SA1210">
+        <context>Using directives must be sorted alphabetically by the namespaces.</context>
+        <line>2</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementMustNotBeOnSingleLine" ruleCheckId="SA1502">
+        <context>The property must not be placed on a single line. The opening and closing curly brackets must each be placed on their own line.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to instance must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>8</line>
+        <index>213</index>
+        <endIndex>220</endIndex>
+        <startLine>8</startLine>
+        <startColumn>53</startColumn>
+        <endLine>8</endLine>
+        <endColumn>60</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="FieldNamesMustBeginWithLowerCaseLetter" ruleCheckId="SA1306">
+        <context>Variable names and private field names must start with a lower-case letter: AIPrefab.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="FieldNamesMustBeginWithLowerCaseLetter" ruleCheckId="SA1306">
+        <context>Variable names and private field names must start with a lower-case letter: PlayerPrefab.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="FieldNamesMustBeginWithLowerCaseLetter" ruleCheckId="SA1306">
+        <context>Variable names and private field names must start with a lower-case letter: Bots.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="FieldNamesMustBeginWithLowerCaseLetter" ruleCheckId="SA1306">
+        <context>Variable names and private field names must start with a lower-case letter: Player.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>18</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>21</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the else statement must be wrapped in opening and closing curly brackets.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to instance must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>21</line>
+        <index>657</index>
+        <endIndex>664</endIndex>
+        <startLine>21</startLine>
+        <startColumn>13</startColumn>
+        <endLine>21</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to instance must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>20</line>
+        <index>626</index>
+        <endIndex>633</endIndex>
+        <startLine>20</startLine>
+        <startColumn>13</startColumn>
+        <endLine>20</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Destroy must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>23</line>
+        <index>701</index>
+        <endIndex>707</endIndex>
+        <startLine>23</startLine>
+        <startColumn>13</startColumn>
+        <endLine>23</endLine>
+        <endColumn>19</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to gameObject must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>23</line>
+        <index>709</index>
+        <endIndex>718</endIndex>
+        <startLine>23</startLine>
+        <startColumn>21</startColumn>
+        <endLine>23</endLine>
+        <endColumn>30</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>26</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the for statement must be wrapped in opening and closing curly brackets.</context>
+        <line>37</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the for statement must be wrapped in opening and closing curly brackets.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>43</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>46</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="ParametersMustBeOnSameLineOrSeparateLines" ruleCheckId="SA1117">
+        <context>All method parameters must be placed on the same line, or each parameter must be placed on a separate line.</context>
+        <line>37</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="SplitParametersMustStartOnLineAfterDeclaration" ruleCheckId="SA1116">
+        <context>If the method parameters are on separate lines, the first parameter must begin on the line beneath the name of the method.</context>
+        <line>37</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="ParameterMustFollowComma" ruleCheckId="SA1115">
+        <context>The parameter must begin on the line after the previous parameter.</context>
+        <line>37</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="ParametersMustBeOnSameLineOrSeparateLines" ruleCheckId="SA1117">
+        <context>All method parameters must be placed on the same line, or each parameter must be placed on a separate line.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="SplitParametersMustStartOnLineAfterDeclaration" ruleCheckId="SA1116">
+        <context>If the method parameters are on separate lines, the first parameter must begin on the line beneath the name of the method.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="ParameterMustFollowComma" ruleCheckId="SA1115">
+        <context>The parameter must begin on the line after the previous parameter.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="ParametersMustBeOnSameLineOrSeparateLines" ruleCheckId="SA1117">
+        <context>All method parameters must be placed on the same line, or each parameter must be placed on a separate line.</context>
+        <line>43</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="SplitParametersMustStartOnLineAfterDeclaration" ruleCheckId="SA1116">
+        <context>If the method parameters are on separate lines, the first parameter must begin on the line beneath the name of the method.</context>
+        <line>43</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="ParameterMustFollowComma" ruleCheckId="SA1115">
+        <context>The parameter must begin on the line after the previous parameter.</context>
+        <line>43</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="ParametersMustBeOnSameLineOrSeparateLines" ruleCheckId="SA1117">
+        <context>All method parameters must be placed on the same line, or each parameter must be placed on a separate line.</context>
+        <line>46</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="SplitParametersMustStartOnLineAfterDeclaration" ruleCheckId="SA1116">
+        <context>If the method parameters are on separate lines, the first parameter must begin on the line beneath the name of the method.</context>
+        <line>46</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="ParameterMustFollowComma" ruleCheckId="SA1115">
+        <context>The parameter must begin on the line after the previous parameter.</context>
+        <line>46</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to InstanciateEntity must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>37</line>
+        <index>1298</index>
+        <endIndex>1314</endIndex>
+        <startLine>37</startLine>
+        <startColumn>13</startColumn>
+        <endLine>37</endLine>
+        <endColumn>29</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForAttackersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>38</line>
+        <index>1366</index>
+        <endIndex>1392</endIndex>
+        <startLine>38</startLine>
+        <startColumn>17</startColumn>
+        <endLine>38</endLine>
+        <endColumn>43</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForAttackersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>38</line>
+        <index>1410</index>
+        <endIndex>1436</endIndex>
+        <startLine>38</startLine>
+        <startColumn>61</startColumn>
+        <endLine>38</endLine>
+        <endColumn>87</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to InstanciateEntity must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>40</line>
+        <index>1514</index>
+        <endIndex>1530</endIndex>
+        <startLine>40</startLine>
+        <startColumn>13</startColumn>
+        <endLine>40</endLine>
+        <endColumn>29</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForDefendersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>41</line>
+        <index>1582</index>
+        <endIndex>1608</endIndex>
+        <startLine>41</startLine>
+        <startColumn>17</startColumn>
+        <endLine>41</endLine>
+        <endColumn>43</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForDefendersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>41</line>
+        <index>1626</index>
+        <endIndex>1652</endIndex>
+        <startLine>41</startLine>
+        <startColumn>61</startColumn>
+        <endLine>41</endLine>
+        <endColumn>87</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to InstanciateEntity must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>43</line>
+        <index>1705</index>
+        <endIndex>1721</endIndex>
+        <startLine>43</startLine>
+        <startColumn>13</startColumn>
+        <endLine>43</endLine>
+        <endColumn>29</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForAttackersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>44</line>
+        <index>1772</index>
+        <endIndex>1798</endIndex>
+        <startLine>44</startLine>
+        <startColumn>17</startColumn>
+        <endLine>44</endLine>
+        <endColumn>43</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForAttackersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>44</line>
+        <index>1816</index>
+        <endIndex>1842</endIndex>
+        <startLine>44</startLine>
+        <startColumn>61</startColumn>
+        <endLine>44</endLine>
+        <endColumn>87</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to InstanciateEntity must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>46</line>
+        <index>1895</index>
+        <endIndex>1911</endIndex>
+        <startLine>46</startLine>
+        <startColumn>13</startColumn>
+        <endLine>46</endLine>
+        <endColumn>29</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForDefendersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>47</line>
+        <index>1962</index>
+        <endIndex>1988</endIndex>
+        <startLine>47</startLine>
+        <startColumn>17</startColumn>
+        <endLine>47</endLine>
+        <endColumn>43</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForDefendersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>47</line>
+        <index>2006</index>
+        <endIndex>2032</endIndex>
+        <startLine>47</startLine>
+        <startColumn>61</startColumn>
+        <endLine>47</endLine>
+        <endColumn>87</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to ResetCharacters must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>49</line>
+        <index>2083</index>
+        <endIndex>2097</endIndex>
+        <startLine>49</startLine>
+        <startColumn>37</startColumn>
+        <endLine>49</endLine>
+        <endColumn>51</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>52</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>60</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the else statement must be wrapped in opening and closing curly brackets.</context>
+        <line>62</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to PlayerPrefab must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>55</line>
+        <index>2289</index>
+        <endIndex>2300</endIndex>
+        <startLine>55</startLine>
+        <startColumn>40</startColumn>
+        <endLine>55</endLine>
+        <endColumn>51</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to AIPrefab must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>55</line>
+        <index>2304</index>
+        <endIndex>2311</endIndex>
+        <startLine>55</startLine>
+        <startColumn>55</startColumn>
+        <endLine>55</endLine>
+        <endColumn>62</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Player must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>67</line>
+        <index>2693</index>
+        <endIndex>2698</endIndex>
+        <startLine>67</startLine>
+        <startColumn>13</startColumn>
+        <endLine>67</endLine>
+        <endColumn>18</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Bots must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>73</line>
+        <index>2922</index>
+        <endIndex>2925</endIndex>
+        <startLine>73</startLine>
+        <startColumn>13</startColumn>
+        <endLine>73</endLine>
+        <endColumn>16</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>77</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>84</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the else statement must be wrapped in opening and closing curly brackets.</context>
+        <line>86</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>93</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the else statement must be wrapped in opening and closing curly brackets.</context>
+        <line>95</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>87</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForAttackersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>84</line>
+        <index>3236</index>
+        <endIndex>3262</endIndex>
+        <startLine>84</startLine>
+        <startColumn>42</startColumn>
+        <endLine>84</endLine>
+        <endColumn>68</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForAttackersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>84</line>
+        <index>3280</index>
+        <endIndex>3306</endIndex>
+        <startLine>84</startLine>
+        <startColumn>86</startColumn>
+        <endLine>84</endLine>
+        <endColumn>112</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForDefendersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>86</line>
+        <index>3386</index>
+        <endIndex>3412</endIndex>
+        <startLine>86</startLine>
+        <startColumn>42</startColumn>
+        <endLine>86</endLine>
+        <endColumn>68</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForDefendersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>86</line>
+        <index>3430</index>
+        <endIndex>3456</endIndex>
+        <startLine>86</startLine>
+        <startColumn>86</startColumn>
+        <endLine>86</endLine>
+        <endColumn>112</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Bots must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>79</line>
+        <index>3036</index>
+        <endIndex>3039</endIndex>
+        <startLine>79</startLine>
+        <startColumn>29</startColumn>
+        <endLine>79</endLine>
+        <endColumn>32</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Player must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>93</line>
+        <index>3688</index>
+        <endIndex>3693</endIndex>
+        <startLine>93</startLine>
+        <startColumn>17</startColumn>
+        <endLine>93</endLine>
+        <endColumn>22</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForAttackersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>93</line>
+        <index>3716</index>
+        <endIndex>3742</endIndex>
+        <startLine>93</startLine>
+        <startColumn>45</startColumn>
+        <endLine>93</endLine>
+        <endColumn>71</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForAttackersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>93</line>
+        <index>3760</index>
+        <endIndex>3786</endIndex>
+        <startLine>93</startLine>
+        <startColumn>89</startColumn>
+        <endLine>93</endLine>
+        <endColumn>115</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Player must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>95</line>
+        <index>3841</index>
+        <endIndex>3846</endIndex>
+        <startLine>95</startLine>
+        <startColumn>17</startColumn>
+        <endLine>95</endLine>
+        <endColumn>22</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForDefendersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>95</line>
+        <index>3869</index>
+        <endIndex>3895</endIndex>
+        <startLine>95</startLine>
+        <startColumn>45</startColumn>
+        <endLine>95</endLine>
+        <endColumn>71</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPointsForDefendersTeam must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>95</line>
+        <index>3913</index>
+        <endIndex>3939</endIndex>
+        <startLine>95</startLine>
+        <startColumn>89</startColumn>
+        <endLine>95</endLine>
+        <endColumn>115</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to TryGetComponent&lt;Player&gt; must begin with the 'this.', 'base.', 'object.' or 'CharacterFactory.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>89</line>
+        <index>3524</index>
+        <endIndex>3546</endIndex>
+        <startLine>89</startLine>
+        <startColumn>13</startColumn>
+        <endLine>89</endLine>
+        <endColumn>35</endColumn>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="TeamEnum.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 15:21:15.026</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The enum must have a documentation header.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="TypeAIEnum.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 15:21:15.027</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The enum must have a documentation header.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>6</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="Character.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.18 10:42:05.218</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>2</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>2</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All classes must be placed after all interfaces.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all constructors.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>6</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>6</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The constructor must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to Condition must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>11</line>
+        <index>229</index>
+        <endIndex>237</endIndex>
+        <startLine>11</startLine>
+        <startColumn>9</startColumn>
+        <endLine>11</endLine>
+        <endColumn>17</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The interface must have a documentation header.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>17</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="CharacterCondition.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.24 01:00:00.568</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CodeMustNotContainMultipleBlankLinesInARow" ruleCheckId="SA1507">
+        <context>The code must not contain multiple blank lines in a row.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The event must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All events must be placed after all fields.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All events must be placed after all fields.</context>
+        <line>35</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All events must be placed after all fields.</context>
+        <line>48</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All events must be placed after all constructors.</context>
+        <line>62</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The event must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The event must have a documentation header.</context>
+        <line>9</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>9</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>17</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to health must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>16</line>
+        <index>342</index>
+        <endIndex>347</endIndex>
+        <startLine>16</startLine>
+        <startColumn>20</startColumn>
+        <endLine>16</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>18</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to health must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>20</line>
+        <index>408</index>
+        <endIndex>413</endIndex>
+        <startLine>20</startLine>
+        <startColumn>13</startColumn>
+        <endLine>20</endLine>
+        <endColumn>18</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to OnChangeHealthEvent must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>21</line>
+        <index>437</index>
+        <endIndex>455</endIndex>
+        <startLine>21</startLine>
+        <startColumn>13</startColumn>
+        <endLine>21</endLine>
+        <endColumn>31</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>25</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>28</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the else statement must be wrapped in opening and closing curly brackets.</context>
+        <line>30</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the else statement must be wrapped in opening and closing curly brackets.</context>
+        <line>32</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the else statement must be wrapped in opening and closing curly brackets.</context>
+        <line>33</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All methods must be placed after all properties.</context>
+        <line>36</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All methods must be placed after all properties.</context>
+        <line>49</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to health must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>27</line>
+        <index>558</index>
+        <endIndex>563</endIndex>
+        <startLine>27</startLine>
+        <startColumn>13</startColumn>
+        <endLine>27</endLine>
+        <endColumn>18</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to health must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>29</line>
+        <index>612</index>
+        <endIndex>617</endIndex>
+        <startLine>29</startLine>
+        <startColumn>18</startColumn>
+        <endLine>29</endLine>
+        <endColumn>23</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to health must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>31</line>
+        <index>666</index>
+        <endIndex>671</endIndex>
+        <startLine>31</startLine>
+        <startColumn>18</startColumn>
+        <endLine>31</endLine>
+        <endColumn>23</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>35</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>35</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>36</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>36</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>47</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>41</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to armour must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>40</line>
+        <index>843</index>
+        <endIndex>848</endIndex>
+        <startLine>40</startLine>
+        <startColumn>20</startColumn>
+        <endLine>40</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>42</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to armour must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>44</line>
+        <index>907</index>
+        <endIndex>912</endIndex>
+        <startLine>44</startLine>
+        <startColumn>13</startColumn>
+        <endLine>44</endLine>
+        <endColumn>18</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to OnChangeArmourEvent must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>45</line>
+        <index>936</index>
+        <endIndex>954</endIndex>
+        <startLine>45</startLine>
+        <startColumn>13</startColumn>
+        <endLine>45</endLine>
+        <endColumn>31</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>48</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>48</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>49</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>49</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>54</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to ammo must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>53</line>
+        <index>1091</index>
+        <endIndex>1094</endIndex>
+        <startLine>53</startLine>
+        <startColumn>20</startColumn>
+        <endLine>53</endLine>
+        <endColumn>23</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>55</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to ammo must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>57</line>
+        <index>1153</index>
+        <endIndex>1156</endIndex>
+        <startLine>57</startLine>
+        <startColumn>13</startColumn>
+        <endLine>57</endLine>
+        <endColumn>16</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to OnChangeAmmunitionEvent must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>58</line>
+        <index>1180</index>
+        <endIndex>1202</endIndex>
+        <startLine>58</startLine>
+        <startColumn>13</startColumn>
+        <endLine>58</endLine>
+        <endColumn>35</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The constructor must have a documentation header.</context>
+        <line>62</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to ammo must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>65</line>
+        <index>1350</index>
+        <endIndex>1353</endIndex>
+        <startLine>65</startLine>
+        <startColumn>9</startColumn>
+        <endLine>65</endLine>
+        <endColumn>12</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to health must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>66</line>
+        <index>1384</index>
+        <endIndex>1389</endIndex>
+        <startLine>66</startLine>
+        <startColumn>9</startColumn>
+        <endLine>66</endLine>
+        <endColumn>14</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to armour must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>67</line>
+        <index>1422</index>
+        <endIndex>1427</endIndex>
+        <startLine>67</startLine>
+        <startColumn>9</startColumn>
+        <endLine>67</endLine>
+        <endColumn>14</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>70</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to HealthPoints must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>70</line>
+        <index>1503</index>
+        <endIndex>1514</endIndex>
+        <startLine>70</startLine>
+        <startColumn>43</startColumn>
+        <endLine>70</endLine>
+        <endColumn>54</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to HealthPoints must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>70</line>
+        <index>1539</index>
+        <endIndex>1550</endIndex>
+        <startLine>70</startLine>
+        <startColumn>79</startColumn>
+        <endLine>70</endLine>
+        <endColumn>90</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>71</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>71</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to ArmourPoints must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>71</line>
+        <index>1605</index>
+        <endIndex>1616</endIndex>
+        <startLine>71</startLine>
+        <startColumn>43</startColumn>
+        <endLine>71</endLine>
+        <endColumn>54</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to ArmourPoints must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>71</line>
+        <index>1641</index>
+        <endIndex>1652</endIndex>
+        <startLine>71</startLine>
+        <startColumn>79</startColumn>
+        <endLine>71</endLine>
+        <endColumn>90</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>72</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>72</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixLocalCallsWithThis" ruleCheckId="SA1101">
+        <context>The call to Ammunition must begin with the 'this.' prefix to indicate that the item is a member of the class.</context>
+        <line>72</line>
+        <index>1703</index>
+        <endIndex>1712</endIndex>
+        <startLine>72</startLine>
+        <startColumn>39</startColumn>
+        <endLine>72</endLine>
+        <endColumn>48</endColumn>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="MovementController.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.28 18:24:37.771</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="UsingDirectivesMustBeOrderedAlphabeticallyByNamespace" ruleCheckId="SA1210">
+        <context>Using directives must be sorted alphabetically by the namespaces.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="SystemUsingDirectivesMustBePlacedBeforeOtherUsingDirectives" ruleCheckId="SA1208">
+        <context>System using directives must be placed before all other using directives.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CodeMustNotContainMultipleBlankLinesInARow" ruleCheckId="SA1507">
+        <context>The code must not contain multiple blank lines in a row.</context>
+        <line>17</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ConstFieldNamesMustBeginWithUpperCaseLetter" ruleCheckId="SA1303">
+        <context>Constants must start with an upper-case letter: updateFlagPositionDelay.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>18</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private methods must be placed after all public methods.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private methods must be placed after all public methods.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private methods must be placed after all public methods.</context>
+        <line>47</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to navMeshAgent must begin with the 'this.', 'base.', 'object.' or 'MovementController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>20</line>
+        <index>555</index>
+        <endIndex>566</endIndex>
+        <startLine>20</startLine>
+        <startColumn>9</startColumn>
+        <endLine>20</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to InvokeRepeating must begin with the 'this.', 'base.', 'object.' or 'MovementController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>21</line>
+        <index>636</index>
+        <endIndex>650</endIndex>
+        <startLine>21</startLine>
+        <startColumn>9</startColumn>
+        <endLine>21</endLine>
+        <endColumn>23</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to UpdateFlagPosition must begin with the 'this.', 'base.', 'object.' or 'MovementController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>21</line>
+        <index>659</index>
+        <endIndex>676</endIndex>
+        <startLine>21</startLine>
+        <startColumn>32</startColumn>
+        <endLine>21</endLine>
+        <endColumn>49</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to CancelInvoke must begin with the 'this.', 'base.', 'object.' or 'MovementController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>26</line>
+        <index>764</index>
+        <endIndex>775</endIndex>
+        <startLine>26</startLine>
+        <startColumn>9</startColumn>
+        <endLine>26</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to UpdateFlagPosition must begin with the 'this.', 'base.', 'object.' or 'MovementController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>26</line>
+        <index>784</index>
+        <endIndex>801</endIndex>
+        <startLine>26</startLine>
+        <startColumn>29</startColumn>
+        <endLine>26</endLine>
+        <endColumn>46</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>29</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to FlagDistance must begin with the 'this.', 'base.', 'object.' or 'MovementController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>31</line>
+        <index>870</index>
+        <endIndex>881</endIndex>
+        <startLine>31</startLine>
+        <startColumn>9</startColumn>
+        <endLine>31</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to flag must begin with the 'this.', 'base.', 'object.' or 'MovementController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>31</line>
+        <index>886</index>
+        <endIndex>889</endIndex>
+        <startLine>31</startLine>
+        <startColumn>25</startColumn>
+        <endLine>31</endLine>
+        <endColumn>28</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to goToNextNavPoint must begin with the 'this.', 'base.', 'object.' or 'MovementController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>37</line>
+        <index>1069</index>
+        <endIndex>1084</endIndex>
+        <startLine>37</startLine>
+        <startColumn>9</startColumn>
+        <endLine>37</endLine>
+        <endColumn>24</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>method names begin with an upper-case letter: getPointsCandidate.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>47</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>method names begin with an upper-case letter: goToNextNavPoint.</context>
+        <line>47</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to navMeshAgent must begin with the 'this.', 'base.', 'object.' or 'MovementController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>48</line>
+        <index>1492</index>
+        <endIndex>1503</endIndex>
+        <startLine>48</startLine>
+        <startColumn>9</startColumn>
+        <endLine>48</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="NPC_State.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.18 09:03:31.781</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The enum must have a documentation header.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The interface must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="InterfaceNamesMustBeginWithI" ruleCheckId="SA1302">
+        <context>Interface names must start with the capital letter 'I': NPC_BaseState.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>18</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>20</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>21</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>21</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>22</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>22</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>25</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>25</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>28</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FileMayOnlyContainASingleClass" ruleCheckId="SA1402">
+        <context>A C# document may only contain a single class at the root level unless all of the classes are partial and are of the same type.</context>
+        <line>28</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>30</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>31</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>31</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>32</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>32</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>33</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>33</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>35</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>35</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>38</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>41</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>41</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>42</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>42</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>43</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>43</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>44</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>44</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>45</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>45</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="Player.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.18 10:43:46.328</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all events.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to PlayerCharacter must begin with the 'this.', 'base.', 'object.' or 'Player.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>10</line>
+        <index>232</index>
+        <endIndex>246</endIndex>
+        <startLine>10</startLine>
+        <startColumn>38</startColumn>
+        <endLine>10</endLine>
+        <endColumn>52</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private methods must be placed after all public methods.</context>
+        <line>18</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private methods must be placed after all public methods.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to PlayerCharacter must begin with the 'this.', 'base.', 'object.' or 'Player.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>14</line>
+        <index>293</index>
+        <endIndex>307</endIndex>
+        <startLine>14</startLine>
+        <startColumn>9</startColumn>
+        <endLine>14</endLine>
+        <endColumn>23</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Condition must begin with the 'this.', 'base.', 'object.' or 'Player.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>15</line>
+        <index>337</index>
+        <endIndex>345</endIndex>
+        <startLine>15</startLine>
+        <startColumn>9</startColumn>
+        <endLine>15</endLine>
+        <endColumn>17</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to PlayerCharacter must begin with the 'this.', 'base.', 'object.' or 'Player.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>15</line>
+        <index>349</index>
+        <endIndex>363</endIndex>
+        <startLine>15</startLine>
+        <startColumn>21</startColumn>
+        <endLine>15</endLine>
+        <endColumn>35</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>18</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Condition must begin with the 'this.', 'base.', 'object.' or 'Player.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>20</line>
+        <index>435</index>
+        <endIndex>443</endIndex>
+        <startLine>20</startLine>
+        <startColumn>9</startColumn>
+        <endLine>20</endLine>
+        <endColumn>17</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The event must have a documentation header.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>31</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="ArithmeticExpressionsMustDeclarePrecedence" ruleCheckId="SA1407">
+        <context>Insert parentheses within the arithmetic expression to declare the operator precedence.</context>
+        <line>27</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to PlayerCharacter must begin with the 'this.', 'base.', 'object.' or 'Player.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>26</line>
+        <index>587</index>
+        <endIndex>601</endIndex>
+        <startLine>26</startLine>
+        <startColumn>9</startColumn>
+        <endLine>26</endLine>
+        <endColumn>23</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Condition must begin with the 'this.', 'base.', 'object.' or 'Player.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>27</line>
+        <index>660</index>
+        <endIndex>668</endIndex>
+        <startLine>27</startLine>
+        <startColumn>9</startColumn>
+        <endLine>27</endLine>
+        <endColumn>17</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Condition must begin with the 'this.', 'base.', 'object.' or 'Player.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>27</line>
+        <index>713</index>
+        <endIndex>721</endIndex>
+        <startLine>27</startLine>
+        <startColumn>62</startColumn>
+        <endLine>27</endLine>
+        <endColumn>70</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Condition must begin with the 'this.', 'base.', 'object.' or 'Player.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>28</line>
+        <index>756</index>
+        <endIndex>764</endIndex>
+        <startLine>28</startLine>
+        <startColumn>9</startColumn>
+        <endLine>28</endLine>
+        <endColumn>17</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to OnKilledEvent must begin with the 'this.', 'base.', 'object.' or 'Player.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>31</line>
+        <index>884</index>
+        <endIndex>896</endIndex>
+        <startLine>31</startLine>
+        <startColumn>13</startColumn>
+        <endLine>31</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Condition must begin with the 'this.', 'base.', 'object.' or 'Player.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>30</line>
+        <index>836</index>
+        <endIndex>844</endIndex>
+        <startLine>30</startLine>
+        <startColumn>13</startColumn>
+        <endLine>30</endLine>
+        <endColumn>21</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="scr_CharacterController.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 15:21:15.030</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>9</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="OpeningCurlyBracketsMustNotBeFollowedByBlankLine" ruleCheckId="SA1505">
+        <context>An opening curly bracket must not be followed by a blank line.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketsMustNotBePrecededByBlankLine" ruleCheckId="SA1508">
+        <context>A closing curly bracket must not be preceded by a blank line.</context>
+        <line>259</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>class names begin with an upper-case letter: scr_CharacterController.</context>
+        <line>9</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>22</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>26</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>28</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>29</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>30</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>33</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>37</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>41</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>42</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>43</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>44</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>45</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private fields must be placed after all public fields.</context>
+        <line>56</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="FieldNamesMustNotContainUnderscore" ruleCheckId="SA1310">
+        <context>Field names must not contain underscores.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="FieldNamesMustNotContainUnderscore" ruleCheckId="SA1310">
+        <context>Field names must not contain underscores.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: input_View.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>18</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>19</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>22</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>22</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: cameraHolder.</context>
+        <line>22</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: feetTransform.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>26</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>26</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: playerSettings.</context>
+        <line>26</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>28</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>28</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>29</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>29</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>30</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>30</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: playerMask.</context>
+        <line>30</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>33</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>33</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: gravityAmount.</context>
+        <line>33</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: gravityMin.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>35</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>37</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>37</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: jumpingForce.</context>
+        <line>37</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>38</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>41</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>41</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: playerStance.</context>
+        <line>41</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>42</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>42</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: playerStanceSmoothing.</context>
+        <line>42</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>43</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>43</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: playerStandStance.</context>
+        <line>43</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>44</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>44</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: playerCrouchStance.</context>
+        <line>44</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>45</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>45</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: playerProneStance.</context>
+        <line>45</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>46</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>48</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>49</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>51</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>53</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>54</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>56</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>56</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="AccessibleFieldsMustBeginWithUpperCaseLetter" ruleCheckId="SA1307">
+        <context>Public and internal fields must start with an upper-case letter: currentWeapon.</context>
+        <line>56</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>57</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>57</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketsMustNotBePrecededByBlankLine" ruleCheckId="SA1508">
+        <context>A closing curly bracket must not be preceded by a blank line.</context>
+        <line>84</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to defaultInput must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>59</line>
+        <index>1597</index>
+        <endIndex>1608</endIndex>
+        <startLine>59</startLine>
+        <startColumn>9</startColumn>
+        <endLine>59</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to defaultInput must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>61</line>
+        <index>1643</index>
+        <endIndex>1654</endIndex>
+        <startLine>61</startLine>
+        <startColumn>9</startColumn>
+        <endLine>61</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to input_Movement must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>61</line>
+        <index>1693</index>
+        <endIndex>1706</endIndex>
+        <startLine>61</startLine>
+        <startColumn>59</startColumn>
+        <endLine>61</endLine>
+        <endColumn>72</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to defaultInput must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>62</line>
+        <index>1743</index>
+        <endIndex>1754</endIndex>
+        <startLine>62</startLine>
+        <startColumn>9</startColumn>
+        <endLine>62</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to input_View must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>62</line>
+        <index>1789</index>
+        <endIndex>1798</endIndex>
+        <startLine>62</startLine>
+        <startColumn>55</startColumn>
+        <endLine>62</endLine>
+        <endColumn>64</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to defaultInput must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>63</line>
+        <index>1835</index>
+        <endIndex>1846</endIndex>
+        <startLine>63</startLine>
+        <startColumn>9</startColumn>
+        <endLine>63</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Jump must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>63</line>
+        <index>1881</index>
+        <endIndex>1884</endIndex>
+        <startLine>63</startLine>
+        <startColumn>55</startColumn>
+        <endLine>63</endLine>
+        <endColumn>58</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to defaultInput must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>65</line>
+        <index>1908</index>
+        <endIndex>1919</endIndex>
+        <startLine>65</startLine>
+        <startColumn>9</startColumn>
+        <endLine>65</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Crouch must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>65</line>
+        <index>1956</index>
+        <endIndex>1961</endIndex>
+        <startLine>65</startLine>
+        <startColumn>57</startColumn>
+        <endLine>65</endLine>
+        <endColumn>62</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to defaultInput must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>66</line>
+        <index>1975</index>
+        <endIndex>1986</endIndex>
+        <startLine>66</startLine>
+        <startColumn>9</startColumn>
+        <endLine>66</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Prone must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>66</line>
+        <index>2022</index>
+        <endIndex>2026</endIndex>
+        <startLine>66</startLine>
+        <startColumn>56</startColumn>
+        <endLine>66</endLine>
+        <endColumn>60</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to defaultInput must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>68</line>
+        <index>2050</index>
+        <endIndex>2061</endIndex>
+        <startLine>68</startLine>
+        <startColumn>9</startColumn>
+        <endLine>68</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to ToggleSprint must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>68</line>
+        <index>2098</index>
+        <endIndex>2109</endIndex>
+        <startLine>68</startLine>
+        <startColumn>57</startColumn>
+        <endLine>68</endLine>
+        <endColumn>68</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to defaultInput must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>69</line>
+        <index>2123</index>
+        <endIndex>2134</endIndex>
+        <startLine>69</startLine>
+        <startColumn>9</startColumn>
+        <endLine>69</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to StopSprint must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>69</line>
+        <index>2179</index>
+        <endIndex>2188</endIndex>
+        <startLine>69</startLine>
+        <startColumn>65</startColumn>
+        <endLine>69</endLine>
+        <endColumn>74</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to defaultInput must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>71</line>
+        <index>2210</index>
+        <endIndex>2221</endIndex>
+        <startLine>71</startLine>
+        <startColumn>9</startColumn>
+        <endLine>71</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to newCameraRotation must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>73</line>
+        <index>2244</index>
+        <endIndex>2260</endIndex>
+        <startLine>73</startLine>
+        <startColumn>9</startColumn>
+        <endLine>73</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to cameraHolder must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>73</line>
+        <index>2264</index>
+        <endIndex>2275</endIndex>
+        <startLine>73</startLine>
+        <startColumn>29</startColumn>
+        <endLine>73</endLine>
+        <endColumn>40</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to newCharacterRotation must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>74</line>
+        <index>2313</index>
+        <endIndex>2332</endIndex>
+        <startLine>74</startLine>
+        <startColumn>9</startColumn>
+        <endLine>74</endLine>
+        <endColumn>28</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to characterController must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>75</line>
+        <index>2382</index>
+        <endIndex>2400</endIndex>
+        <startLine>75</startLine>
+        <startColumn>9</startColumn>
+        <endLine>75</endLine>
+        <endColumn>27</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to GetComponent&lt;CharacterController&gt; must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>75</line>
+        <index>2404</index>
+        <endIndex>2436</endIndex>
+        <startLine>75</startLine>
+        <startColumn>31</startColumn>
+        <endLine>75</endLine>
+        <endColumn>63</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to cameraHeight must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>77</line>
+        <index>2452</index>
+        <endIndex>2463</endIndex>
+        <startLine>77</startLine>
+        <startColumn>9</startColumn>
+        <endLine>77</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to cameraHolder must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>77</line>
+        <index>2467</index>
+        <endIndex>2478</endIndex>
+        <startLine>77</startLine>
+        <startColumn>24</startColumn>
+        <endLine>77</endLine>
+        <endColumn>35</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to currentWeapon must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>81</line>
+        <index>2551</index>
+        <endIndex>2563</endIndex>
+        <startLine>81</startLine>
+        <startColumn>13</startColumn>
+        <endLine>81</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to currentWeapon must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>79</line>
+        <index>2512</index>
+        <endIndex>2524</endIndex>
+        <startLine>79</startLine>
+        <startColumn>13</startColumn>
+        <endLine>79</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>86</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to CalculateView must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>88</line>
+        <index>2656</index>
+        <endIndex>2668</endIndex>
+        <startLine>88</startLine>
+        <startColumn>9</startColumn>
+        <endLine>88</endLine>
+        <endColumn>21</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to CalculateMovement must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>89</line>
+        <index>2682</index>
+        <endIndex>2698</endIndex>
+        <startLine>89</startLine>
+        <startColumn>9</startColumn>
+        <endLine>89</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to CalculateJump must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>90</line>
+        <index>2712</index>
+        <endIndex>2724</endIndex>
+        <startLine>90</startLine>
+        <startColumn>9</startColumn>
+        <endLine>90</endLine>
+        <endColumn>21</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to CalculateCameraHeight must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>91</line>
+        <index>2738</index>
+        <endIndex>2758</endIndex>
+        <startLine>91</startLine>
+        <startColumn>9</startColumn>
+        <endLine>91</endLine>
+        <endColumn>29</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>94</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to newCharacterRotation must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>96</line>
+        <index>2822</index>
+        <endIndex>2841</endIndex>
+        <startLine>96</startLine>
+        <startColumn>9</startColumn>
+        <endLine>96</endLine>
+        <endColumn>28</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>96</line>
+        <index>2848</index>
+        <endIndex>2861</endIndex>
+        <startLine>96</startLine>
+        <startColumn>35</startColumn>
+        <endLine>96</endLine>
+        <endColumn>48</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>96</line>
+        <index>2883</index>
+        <endIndex>2896</endIndex>
+        <startLine>96</startLine>
+        <startColumn>70</startColumn>
+        <endLine>96</endLine>
+        <endColumn>83</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to input_View must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>96</line>
+        <index>2915</index>
+        <endIndex>2924</endIndex>
+        <startLine>96</startLine>
+        <startColumn>102</startColumn>
+        <endLine>96</endLine>
+        <endColumn>111</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to input_View must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>96</line>
+        <index>2930</index>
+        <endIndex>2939</endIndex>
+        <startLine>96</startLine>
+        <startColumn>117</startColumn>
+        <endLine>96</endLine>
+        <endColumn>126</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to newCharacterRotation must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>97</line>
+        <index>3014</index>
+        <endIndex>3033</endIndex>
+        <startLine>97</startLine>
+        <startColumn>52</startColumn>
+        <endLine>97</endLine>
+        <endColumn>71</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to newCameraRotation must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>99</line>
+        <index>3056</index>
+        <endIndex>3072</endIndex>
+        <startLine>99</startLine>
+        <startColumn>9</startColumn>
+        <endLine>99</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>99</line>
+        <index>3079</index>
+        <endIndex>3092</endIndex>
+        <startLine>99</startLine>
+        <startColumn>32</startColumn>
+        <endLine>99</endLine>
+        <endColumn>45</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>99</line>
+        <index>3114</index>
+        <endIndex>3127</endIndex>
+        <startLine>99</startLine>
+        <startColumn>67</startColumn>
+        <endLine>99</endLine>
+        <endColumn>80</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to input_View must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>99</line>
+        <index>3145</index>
+        <endIndex>3154</endIndex>
+        <startLine>99</startLine>
+        <startColumn>98</startColumn>
+        <endLine>99</endLine>
+        <endColumn>107</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to input_View must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>99</line>
+        <index>3161</index>
+        <endIndex>3170</endIndex>
+        <startLine>99</startLine>
+        <startColumn>114</startColumn>
+        <endLine>99</endLine>
+        <endColumn>123</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to newCameraRotation must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>100</line>
+        <index>3202</index>
+        <endIndex>3218</endIndex>
+        <startLine>100</startLine>
+        <startColumn>9</startColumn>
+        <endLine>100</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to newCameraRotation must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>100</line>
+        <index>3236</index>
+        <endIndex>3252</endIndex>
+        <startLine>100</startLine>
+        <startColumn>43</startColumn>
+        <endLine>100</endLine>
+        <endColumn>59</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to ViewClampYMin must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>100</line>
+        <index>3257</index>
+        <endIndex>3269</endIndex>
+        <startLine>100</startLine>
+        <startColumn>64</startColumn>
+        <endLine>100</endLine>
+        <endColumn>76</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to ViewClampYMax must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>100</line>
+        <index>3272</index>
+        <endIndex>3284</endIndex>
+        <startLine>100</startLine>
+        <startColumn>79</startColumn>
+        <endLine>100</endLine>
+        <endColumn>91</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to cameraHolder must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>102</line>
+        <index>3307</index>
+        <endIndex>3318</endIndex>
+        <startLine>102</startLine>
+        <startColumn>9</startColumn>
+        <endLine>102</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to newCameraRotation must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>102</line>
+        <index>3353</index>
+        <endIndex>3369</endIndex>
+        <startLine>102</startLine>
+        <startColumn>55</startColumn>
+        <endLine>102</endLine>
+        <endColumn>71</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>105</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="FieldNamesMustBeginWithLowerCaseLetter" ruleCheckId="SA1306">
+        <context>Variable names and private field names must start with a lower-case letter: MovementSpeed.</context>
+        <line>147</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="ParametersMustBeOnSameLineOrSeparateLines" ruleCheckId="SA1117">
+        <context>All method parameters must be placed on the same line, or each parameter must be placed on a separate line.</context>
+        <line>142</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="SplitParametersMustStartOnLineAfterDeclaration" ruleCheckId="SA1116">
+        <context>If the method parameters are on separate lines, the first parameter must begin on the line beneath the name of the method.</context>
+        <line>142</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="ParameterMustFollowComma" ruleCheckId="SA1115">
+        <context>The parameter must begin on the line after the previous parameter.</context>
+        <line>145</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="ParametersMustBeOnSameLineOrSeparateLines" ruleCheckId="SA1117">
+        <context>All method parameters must be placed on the same line, or each parameter must be placed on a separate line.</context>
+        <line>143</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="SplitParametersMustStartOnLineAfterDeclaration" ruleCheckId="SA1116">
+        <context>If the method parameters are on separate lines, the first parameter must begin on the line beneath the name of the method.</context>
+        <line>143</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="ParameterMustFollowComma" ruleCheckId="SA1115">
+        <context>The parameter must begin on the line after the previous parameter.</context>
+        <line>144</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to isSprinting must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>109</line>
+        <index>3490</index>
+        <endIndex>3500</endIndex>
+        <startLine>109</startLine>
+        <startColumn>13</startColumn>
+        <endLine>109</endLine>
+        <endColumn>23</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to input_Movement must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>107</line>
+        <index>3440</index>
+        <endIndex>3453</endIndex>
+        <startLine>107</startLine>
+        <startColumn>13</startColumn>
+        <endLine>107</endLine>
+        <endColumn>26</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>112</line>
+        <index>3561</index>
+        <endIndex>3574</endIndex>
+        <startLine>112</startLine>
+        <startColumn>29</startColumn>
+        <endLine>112</endLine>
+        <endColumn>42</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>113</line>
+        <index>3628</index>
+        <endIndex>3641</endIndex>
+        <startLine>113</startLine>
+        <startColumn>31</startColumn>
+        <endLine>113</endLine>
+        <endColumn>44</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>117</line>
+        <index>3731</index>
+        <endIndex>3744</endIndex>
+        <startLine>117</startLine>
+        <startColumn>29</startColumn>
+        <endLine>117</endLine>
+        <endColumn>42</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>118</line>
+        <index>3798</index>
+        <endIndex>3811</endIndex>
+        <startLine>118</startLine>
+        <startColumn>31</startColumn>
+        <endLine>118</endLine>
+        <endColumn>44</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to isSprinting must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>115</line>
+        <index>3678</index>
+        <endIndex>3688</endIndex>
+        <startLine>115</startLine>
+        <startColumn>13</startColumn>
+        <endLine>115</endLine>
+        <endColumn>23</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>124</line>
+        <index>3946</index>
+        <endIndex>3959</endIndex>
+        <startLine>124</startLine>
+        <startColumn>13</startColumn>
+        <endLine>124</endLine>
+        <endColumn>26</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>124</line>
+        <index>3977</index>
+        <endIndex>3990</endIndex>
+        <startLine>124</startLine>
+        <startColumn>44</startColumn>
+        <endLine>124</endLine>
+        <endColumn>57</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to characterController must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>122</line>
+        <index>3890</index>
+        <endIndex>3908</endIndex>
+        <startLine>122</startLine>
+        <startColumn>14</startColumn>
+        <endLine>122</endLine>
+        <endColumn>32</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>128</line>
+        <index>4103</index>
+        <endIndex>4116</endIndex>
+        <startLine>128</startLine>
+        <startColumn>13</startColumn>
+        <endLine>128</endLine>
+        <endColumn>26</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>128</line>
+        <index>4134</index>
+        <endIndex>4147</endIndex>
+        <startLine>128</startLine>
+        <startColumn>44</startColumn>
+        <endLine>128</endLine>
+        <endColumn>57</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>126</line>
+        <index>4042</index>
+        <endIndex>4053</endIndex>
+        <startLine>126</startLine>
+        <startColumn>17</startColumn>
+        <endLine>126</endLine>
+        <endColumn>28</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>132</line>
+        <index>4259</index>
+        <endIndex>4272</endIndex>
+        <startLine>132</startLine>
+        <startColumn>13</startColumn>
+        <endLine>132</endLine>
+        <endColumn>26</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>132</line>
+        <index>4290</index>
+        <endIndex>4303</endIndex>
+        <startLine>132</startLine>
+        <startColumn>44</startColumn>
+        <endLine>132</endLine>
+        <endColumn>57</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>130</line>
+        <index>4199</index>
+        <endIndex>4210</endIndex>
+        <startLine>130</startLine>
+        <startColumn>17</startColumn>
+        <endLine>130</endLine>
+        <endColumn>28</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>136</line>
+        <index>4374</index>
+        <endIndex>4387</endIndex>
+        <startLine>136</startLine>
+        <startColumn>13</startColumn>
+        <endLine>136</endLine>
+        <endColumn>26</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>139</line>
+        <index>4455</index>
+        <endIndex>4468</endIndex>
+        <startLine>139</startLine>
+        <startColumn>26</startColumn>
+        <endLine>139</endLine>
+        <endColumn>39</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>140</line>
+        <index>4513</index>
+        <endIndex>4526</endIndex>
+        <startLine>140</startLine>
+        <startColumn>28</startColumn>
+        <endLine>140</endLine>
+        <endColumn>41</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to newMovementSpeed must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>142</line>
+        <index>4562</index>
+        <endIndex>4577</endIndex>
+        <startLine>142</startLine>
+        <startColumn>9</startColumn>
+        <endLine>142</endLine>
+        <endColumn>24</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to newMovementSpeed must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>142</line>
+        <index>4600</index>
+        <endIndex>4615</endIndex>
+        <startLine>142</startLine>
+        <startColumn>47</startColumn>
+        <endLine>142</endLine>
+        <endColumn>62</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to input_Movement must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>143</line>
+        <index>4656</index>
+        <endIndex>4669</endIndex>
+        <startLine>143</startLine>
+        <startColumn>38</startColumn>
+        <endLine>143</endLine>
+        <endColumn>51</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to input_Movement must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>144</line>
+        <index>4721</index>
+        <endIndex>4734</endIndex>
+        <startLine>144</startLine>
+        <startColumn>30</startColumn>
+        <endLine>144</endLine>
+        <endColumn>43</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to newMovementSpeedVelocity must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>145</line>
+        <index>4774</index>
+        <endIndex>4797</endIndex>
+        <startLine>145</startLine>
+        <startColumn>17</startColumn>
+        <endLine>145</endLine>
+        <endColumn>40</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to characterController must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>145</line>
+        <index>4800</index>
+        <endIndex>4818</endIndex>
+        <startLine>145</startLine>
+        <startColumn>43</startColumn>
+        <endLine>145</endLine>
+        <endColumn>61</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>145</line>
+        <index>4833</index>
+        <endIndex>4846</endIndex>
+        <startLine>145</startLine>
+        <startColumn>76</startColumn>
+        <endLine>145</endLine>
+        <endColumn>89</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>145</line>
+        <index>4868</index>
+        <endIndex>4881</endIndex>
+        <startLine>145</startLine>
+        <startColumn>111</startColumn>
+        <endLine>145</endLine>
+        <endColumn>124</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to newMovementSpeed must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>147</line>
+        <index>4970</index>
+        <endIndex>4985</endIndex>
+        <startLine>147</startLine>
+        <startColumn>58</startColumn>
+        <endLine>147</endLine>
+        <endColumn>73</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerGravity must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>151</line>
+        <index>5056</index>
+        <endIndex>5068</endIndex>
+        <startLine>151</startLine>
+        <startColumn>13</startColumn>
+        <endLine>151</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to gravityAmount must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>151</line>
+        <index>5073</index>
+        <endIndex>5085</endIndex>
+        <startLine>151</startLine>
+        <startColumn>30</startColumn>
+        <endLine>151</endLine>
+        <endColumn>42</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerGravity must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>149</line>
+        <index>5004</index>
+        <endIndex>5016</endIndex>
+        <startLine>149</startLine>
+        <startColumn>13</startColumn>
+        <endLine>149</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to gravityMin must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>149</line>
+        <index>5020</index>
+        <endIndex>5029</endIndex>
+        <startLine>149</startLine>
+        <startColumn>29</startColumn>
+        <endLine>149</endLine>
+        <endColumn>38</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerGravity must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>156</line>
+        <index>5212</index>
+        <endIndex>5224</endIndex>
+        <startLine>156</startLine>
+        <startColumn>13</startColumn>
+        <endLine>156</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerGravity must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>154</line>
+        <index>5131</index>
+        <endIndex>5143</endIndex>
+        <startLine>154</startLine>
+        <startColumn>13</startColumn>
+        <endLine>154</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to characterController must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>154</line>
+        <index>5156</index>
+        <endIndex>5174</endIndex>
+        <startLine>154</startLine>
+        <startColumn>38</startColumn>
+        <endLine>154</endLine>
+        <endColumn>56</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerGravity must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>159</line>
+        <index>5276</index>
+        <endIndex>5288</endIndex>
+        <startLine>159</startLine>
+        <startColumn>28</startColumn>
+        <endLine>159</endLine>
+        <endColumn>40</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to jumpingForce must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>160</line>
+        <index>5317</index>
+        <endIndex>5328</endIndex>
+        <startLine>160</startLine>
+        <startColumn>26</startColumn>
+        <endLine>160</endLine>
+        <endColumn>37</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to characterController must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>162</line>
+        <index>5367</index>
+        <endIndex>5385</endIndex>
+        <startLine>162</startLine>
+        <startColumn>9</startColumn>
+        <endLine>162</endLine>
+        <endColumn>27</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="KeywordsMustBeSpacedCorrectly" ruleCheckId="SA1000">
+        <context>The spacing around the keyword 'if' is invalid.</context>
+        <line>126</line>
+        <index>4039</index>
+        <endIndex>4040</endIndex>
+        <startLine>126</startLine>
+        <startColumn>14</startColumn>
+        <endLine>126</endLine>
+        <endColumn>15</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="KeywordsMustBeSpacedCorrectly" ruleCheckId="SA1000">
+        <context>The spacing around the keyword 'if' is invalid.</context>
+        <line>130</line>
+        <index>4196</index>
+        <endIndex>4197</endIndex>
+        <startLine>130</startLine>
+        <startColumn>14</startColumn>
+        <endLine>130</endLine>
+        <endColumn>15</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>165</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to jumpingForce must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>167</line>
+        <index>5467</index>
+        <endIndex>5478</endIndex>
+        <startLine>167</startLine>
+        <startColumn>9</startColumn>
+        <endLine>167</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to jumpingForce must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>167</line>
+        <index>5501</index>
+        <endIndex>5512</endIndex>
+        <startLine>167</startLine>
+        <startColumn>43</startColumn>
+        <endLine>167</endLine>
+        <endColumn>54</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to jumpingForceVelocity must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>167</line>
+        <index>5533</index>
+        <endIndex>5552</endIndex>
+        <startLine>167</startLine>
+        <startColumn>75</startColumn>
+        <endLine>167</endLine>
+        <endColumn>94</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>167</line>
+        <index>5555</index>
+        <endIndex>5568</endIndex>
+        <startLine>167</startLine>
+        <startColumn>97</startColumn>
+        <endLine>167</endLine>
+        <endColumn>110</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>170</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>186</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStandStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>172</line>
+        <index>5673</index>
+        <endIndex>5689</endIndex>
+        <startLine>172</startLine>
+        <startColumn>28</startColumn>
+        <endLine>172</endLine>
+        <endColumn>44</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerCrouchStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>176</line>
+        <index>5796</index>
+        <endIndex>5813</endIndex>
+        <startLine>176</startLine>
+        <startColumn>28</startColumn>
+        <endLine>176</endLine>
+        <endColumn>45</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>174</line>
+        <index>5720</index>
+        <endIndex>5731</endIndex>
+        <startLine>174</startLine>
+        <startColumn>13</startColumn>
+        <endLine>174</endLine>
+        <endColumn>24</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerProneStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>180</line>
+        <index>5933</index>
+        <endIndex>5949</endIndex>
+        <startLine>180</startLine>
+        <startColumn>28</startColumn>
+        <endLine>180</endLine>
+        <endColumn>44</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>178</line>
+        <index>5858</index>
+        <endIndex>5869</endIndex>
+        <startLine>178</startLine>
+        <startColumn>18</startColumn>
+        <endLine>178</endLine>
+        <endColumn>29</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to cameraHeight must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>183</line>
+        <index>5995</index>
+        <endIndex>6006</endIndex>
+        <startLine>183</startLine>
+        <startColumn>9</startColumn>
+        <endLine>183</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to cameraHolder must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>183</line>
+        <index>6027</index>
+        <endIndex>6038</endIndex>
+        <startLine>183</startLine>
+        <startColumn>41</startColumn>
+        <endLine>183</endLine>
+        <endColumn>52</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to cameraHeightVelocity must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>183</line>
+        <index>6075</index>
+        <endIndex>6094</endIndex>
+        <startLine>183</startLine>
+        <startColumn>89</startColumn>
+        <endLine>183</endLine>
+        <endColumn>108</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStanceSmoothing must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>183</line>
+        <index>6097</index>
+        <endIndex>6117</endIndex>
+        <startLine>183</startLine>
+        <startColumn>111</startColumn>
+        <endLine>183</endLine>
+        <endColumn>131</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to cameraHolder must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>185</line>
+        <index>6132</index>
+        <endIndex>6143</endIndex>
+        <startLine>185</startLine>
+        <startColumn>9</startColumn>
+        <endLine>185</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to cameraHolder must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>185</line>
+        <index>6173</index>
+        <endIndex>6184</endIndex>
+        <startLine>185</startLine>
+        <startColumn>50</startColumn>
+        <endLine>185</endLine>
+        <endColumn>61</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to cameraHeight must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>185</line>
+        <index>6203</index>
+        <endIndex>6214</endIndex>
+        <startLine>185</startLine>
+        <startColumn>80</startColumn>
+        <endLine>185</endLine>
+        <endColumn>91</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to cameraHolder must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>185</line>
+        <index>6217</index>
+        <endIndex>6228</endIndex>
+        <startLine>185</startLine>
+        <startColumn>94</startColumn>
+        <endLine>185</endLine>
+        <endColumn>105</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>187</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>187</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>199</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to characterController must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>189</line>
+        <index>6301</index>
+        <endIndex>6319</endIndex>
+        <startLine>189</startLine>
+        <startColumn>14</startColumn>
+        <endLine>189</endLine>
+        <endColumn>32</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>189</line>
+        <index>6335</index>
+        <endIndex>6346</endIndex>
+        <startLine>189</startLine>
+        <startColumn>48</startColumn>
+        <endLine>189</endLine>
+        <endColumn>59</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to StanceCheck must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>196</line>
+        <index>6502</index>
+        <endIndex>6512</endIndex>
+        <startLine>196</startLine>
+        <startColumn>17</startColumn>
+        <endLine>196</endLine>
+        <endColumn>27</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStandStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>196</line>
+        <index>6514</index>
+        <endIndex>6530</endIndex>
+        <startLine>196</startLine>
+        <startColumn>29</startColumn>
+        <endLine>196</endLine>
+        <endColumn>45</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>200</line>
+        <index>6624</index>
+        <endIndex>6635</endIndex>
+        <startLine>200</startLine>
+        <startColumn>13</startColumn>
+        <endLine>200</endLine>
+        <endColumn>24</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>194</line>
+        <index>6437</index>
+        <endIndex>6448</endIndex>
+        <startLine>194</startLine>
+        <startColumn>13</startColumn>
+        <endLine>194</endLine>
+        <endColumn>24</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to jumpingForce must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>205</line>
+        <index>6727</index>
+        <endIndex>6738</endIndex>
+        <startLine>205</startLine>
+        <startColumn>9</startColumn>
+        <endLine>205</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>205</line>
+        <index>6755</index>
+        <endIndex>6768</endIndex>
+        <startLine>205</startLine>
+        <startColumn>37</startColumn>
+        <endLine>205</endLine>
+        <endColumn>50</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerGravity must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>206</line>
+        <index>6794</index>
+        <endIndex>6806</endIndex>
+        <startLine>206</startLine>
+        <startColumn>9</startColumn>
+        <endLine>206</endLine>
+        <endColumn>21</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>209</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>216</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>219</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>223</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to StanceCheck must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>213</line>
+        <index>6934</index>
+        <endIndex>6944</endIndex>
+        <startLine>213</startLine>
+        <startColumn>17</startColumn>
+        <endLine>213</endLine>
+        <endColumn>27</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStandStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>213</line>
+        <index>6946</index>
+        <endIndex>6962</endIndex>
+        <startLine>213</startLine>
+        <startColumn>29</startColumn>
+        <endLine>213</endLine>
+        <endColumn>45</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>217</line>
+        <index>7056</index>
+        <endIndex>7067</endIndex>
+        <startLine>217</startLine>
+        <startColumn>13</startColumn>
+        <endLine>217</endLine>
+        <endColumn>24</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>211</line>
+        <index>6869</index>
+        <endIndex>6880</endIndex>
+        <startLine>211</startLine>
+        <startColumn>13</startColumn>
+        <endLine>211</endLine>
+        <endColumn>24</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to StanceCheck must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>220</line>
+        <index>7136</index>
+        <endIndex>7146</endIndex>
+        <startLine>220</startLine>
+        <startColumn>13</startColumn>
+        <endLine>220</endLine>
+        <endColumn>23</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerCrouchStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>220</line>
+        <index>7148</index>
+        <endIndex>7165</endIndex>
+        <startLine>220</startLine>
+        <startColumn>25</startColumn>
+        <endLine>220</endLine>
+        <endColumn>42</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>224</line>
+        <index>7243</index>
+        <endIndex>7254</endIndex>
+        <startLine>224</startLine>
+        <startColumn>9</startColumn>
+        <endLine>224</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>227</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerStance must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>229</line>
+        <index>7330</index>
+        <endIndex>7341</endIndex>
+        <startLine>229</startLine>
+        <startColumn>9</startColumn>
+        <endLine>229</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>232</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CodeMustNotContainMultipleBlankLinesInARow" ruleCheckId="SA1507">
+        <context>The code must not contain multiple blank lines in a row.</context>
+        <line>237</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to feetTransform must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>234</line>
+        <index>7469</index>
+        <endIndex>7481</endIndex>
+        <startLine>234</startLine>
+        <startColumn>33</startColumn>
+        <endLine>234</endLine>
+        <endColumn>45</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to feetTransform must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>234</line>
+        <index>7495</index>
+        <endIndex>7507</endIndex>
+        <startLine>234</startLine>
+        <startColumn>59</startColumn>
+        <endLine>234</endLine>
+        <endColumn>71</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to characterController must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>234</line>
+        <index>7522</index>
+        <endIndex>7540</endIndex>
+        <startLine>234</startLine>
+        <startColumn>86</startColumn>
+        <endLine>234</endLine>
+        <endColumn>104</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to stanceCheckErrorMargin must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>234</line>
+        <index>7551</index>
+        <endIndex>7572</endIndex>
+        <startLine>234</startLine>
+        <startColumn>115</startColumn>
+        <endLine>234</endLine>
+        <endColumn>136</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to feetTransform must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>234</line>
+        <index>7575</index>
+        <endIndex>7587</endIndex>
+        <startLine>234</startLine>
+        <startColumn>139</startColumn>
+        <endLine>234</endLine>
+        <endColumn>151</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to feetTransform must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>235</line>
+        <index>7633</index>
+        <endIndex>7645</endIndex>
+        <startLine>235</startLine>
+        <startColumn>31</startColumn>
+        <endLine>235</endLine>
+        <endColumn>43</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to feetTransform must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>235</line>
+        <index>7659</index>
+        <endIndex>7671</endIndex>
+        <startLine>235</startLine>
+        <startColumn>57</startColumn>
+        <endLine>235</endLine>
+        <endColumn>69</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to characterController must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>235</line>
+        <index>7686</index>
+        <endIndex>7704</endIndex>
+        <startLine>235</startLine>
+        <startColumn>84</startColumn>
+        <endLine>235</endLine>
+        <endColumn>102</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to stanceCheckErrorMargin must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>235</line>
+        <index>7715</index>
+        <endIndex>7736</endIndex>
+        <startLine>235</startLine>
+        <startColumn>113</startColumn>
+        <endLine>235</endLine>
+        <endColumn>134</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to feetTransform must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>235</line>
+        <index>7759</index>
+        <endIndex>7771</endIndex>
+        <startLine>235</startLine>
+        <startColumn>157</startColumn>
+        <endLine>235</endLine>
+        <endColumn>169</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to characterController must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>238</line>
+        <index>7855</index>
+        <endIndex>7873</endIndex>
+        <startLine>238</startLine>
+        <startColumn>49</startColumn>
+        <endLine>238</endLine>
+        <endColumn>67</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerMask must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>238</line>
+        <index>7883</index>
+        <endIndex>7892</endIndex>
+        <startLine>238</startLine>
+        <startColumn>77</startColumn>
+        <endLine>238</endLine>
+        <endColumn>86</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>241</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>247</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to isSprinting must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>245</line>
+        <index>8008</index>
+        <endIndex>8018</endIndex>
+        <startLine>245</startLine>
+        <startColumn>13</startColumn>
+        <endLine>245</endLine>
+        <endColumn>23</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to input_Movement must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>243</line>
+        <index>7958</index>
+        <endIndex>7971</endIndex>
+        <startLine>243</startLine>
+        <startColumn>13</startColumn>
+        <endLine>243</endLine>
+        <endColumn>26</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to isSprinting must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>248</line>
+        <index>8070</index>
+        <endIndex>8080</endIndex>
+        <startLine>248</startLine>
+        <startColumn>9</startColumn>
+        <endLine>248</endLine>
+        <endColumn>19</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to isSprinting must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>248</line>
+        <index>8085</index>
+        <endIndex>8095</endIndex>
+        <startLine>248</startLine>
+        <startColumn>24</startColumn>
+        <endLine>248</endLine>
+        <endColumn>34</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>251</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to isSprinting must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>255</line>
+        <index>8216</index>
+        <endIndex>8226</endIndex>
+        <startLine>255</startLine>
+        <startColumn>13</startColumn>
+        <endLine>255</endLine>
+        <endColumn>23</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to playerSettings must begin with the 'this.', 'base.', 'object.' or 'scr_CharacterController.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>253</line>
+        <index>8162</index>
+        <endIndex>8175</endIndex>
+        <startLine>253</startLine>
+        <startColumn>13</startColumn>
+        <endLine>253</endLine>
+        <endColumn>26</endColumn>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="scr_Models.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 15:21:15.031</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>class names begin with an upper-case letter: scr_Models.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The enum must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>19</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>19</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>20</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>20</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>22</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>22</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>26</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>26</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>27</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>27</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>30</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>30</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>31</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>31</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>35</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>35</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>36</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>36</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>39</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>39</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>41</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>41</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>44</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>44</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>45</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>45</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>46</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>46</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>47</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>47</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>51</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>53</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>53</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>54</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>54</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>62</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>65</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>65</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>66</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>66</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>67</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>67</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>68</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>68</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>69</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>69</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>70</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>70</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>71</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>71</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="GameManager.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.24 01:02:47.068</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="SystemUsingDirectivesMustBePlacedBeforeOtherUsingDirectives" ruleCheckId="SA1208">
+        <context>System using directives must be placed before all other using directives.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementMustNotBeOnSingleLine" ruleCheckId="SA1502">
+        <context>The property must not be placed on a single line. The opening and closing curly brackets must each be placed on their own line.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all events.</context>
+        <line>98</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="FieldNamesMustBeginWithLowerCaseLetter" ruleCheckId="SA1306">
+        <context>Variable names and private field names must start with a lower-case letter: DefendersTeam.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="FieldNamesMustBeginWithLowerCaseLetter" ruleCheckId="SA1306">
+        <context>Variable names and private field names must start with a lower-case letter: AttackersTeam.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the else statement must be wrapped in opening and closing curly brackets.</context>
+        <line>18</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private methods must be placed after all public methods.</context>
+        <line>39</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private methods must be placed after all public methods.</context>
+        <line>54</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustBeOrderedByAccess" ruleCheckId="SA1202">
+        <context>All private methods must be placed after all public methods.</context>
+        <line>68</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Destroy must begin with the 'this.', 'base.', 'object.' or 'GameManager.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>18</line>
+        <index>536</index>
+        <endIndex>542</endIndex>
+        <startLine>18</startLine>
+        <startColumn>13</startColumn>
+        <endLine>18</endLine>
+        <endColumn>19</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to gameObject must begin with the 'this.', 'base.', 'object.' or 'GameManager.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>18</line>
+        <index>544</index>
+        <endIndex>553</endIndex>
+        <startLine>18</startLine>
+        <startColumn>21</startColumn>
+        <endLine>18</endLine>
+        <endColumn>30</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>21</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>33</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the else statement must be wrapped in opening and closing curly brackets.</context>
+        <line>35</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to ResetScene must begin with the 'this.', 'base.', 'object.' or 'GameManager.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>23</line>
+        <index>647</index>
+        <endIndex>656</endIndex>
+        <startLine>23</startLine>
+        <startColumn>48</startColumn>
+        <endLine>23</endLine>
+        <endColumn>57</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to flagCaptured must begin with the 'this.', 'base.', 'object.' or 'GameManager.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>25</line>
+        <index>706</index>
+        <endIndex>717</endIndex>
+        <startLine>25</startLine>
+        <startColumn>45</startColumn>
+        <endLine>25</endLine>
+        <endColumn>56</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to timeOut must begin with the 'this.', 'base.', 'object.' or 'GameManager.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>26</line>
+        <index>762</index>
+        <endIndex>768</endIndex>
+        <startLine>26</startLine>
+        <startColumn>42</startColumn>
+        <endLine>26</endLine>
+        <endColumn>48</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>39</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>43</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the else statement must be wrapped in opening and closing curly brackets.</context>
+        <line>45</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the foreach statement must be wrapped in opening and closing curly brackets.</context>
+        <line>49</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>50</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>54</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>58</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the else statement must be wrapped in opening and closing curly brackets.</context>
+        <line>60</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the foreach statement must be wrapped in opening and closing curly brackets.</context>
+        <line>63</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>64</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>68</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>71</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>method names begin with an upper-case letter: flagCaptured.</context>
+        <line>71</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="KeywordsMustBeSpacedCorrectly" ruleCheckId="SA1000">
+        <context>The spacing around the keyword 'switch' is invalid.</context>
+        <line>73</line>
+        <index>2522</index>
+        <endIndex>2527</endIndex>
+        <startLine>73</startLine>
+        <startColumn>9</startColumn>
+        <endLine>73</endLine>
+        <endColumn>14</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>87</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>method names begin with an upper-case letter: timeOut.</context>
+        <line>87</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>92</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to flagCaptured must begin with the 'this.', 'base.', 'object.' or 'GameManager.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>94</line>
+        <index>3031</index>
+        <endIndex>3042</endIndex>
+        <startLine>94</startLine>
+        <startColumn>45</startColumn>
+        <endLine>94</endLine>
+        <endColumn>56</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to timeOut must begin with the 'this.', 'base.', 'object.' or 'GameManager.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>95</line>
+        <index>3087</index>
+        <endIndex>3093</endIndex>
+        <startLine>95</startLine>
+        <startColumn>42</startColumn>
+        <endLine>95</endLine>
+        <endColumn>48</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The event must have a documentation header.</context>
+        <line>98</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>99</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>99</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="GlobalEventManager.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 15:21:15.033</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The event must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>event names begin with an upper-case letter: onCaptureFlag.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All methods must be placed after all events.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The event must have a documentation header.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>event names begin with an upper-case letter: onTimeLeft.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="MapManager.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.16 19:49:10.120</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>6</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>property names begin with an upper-case letter: navPoints.</context>
+        <line>6</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="CodeMustNotContainMultipleStatementsOnOneLine" ruleCheckId="SA1107">
+        <context>A line may only contain a single statement.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="TimeManager.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.16 00:20:04.403</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementMustNotBeOnSingleLine" ruleCheckId="SA1502">
+        <context>The property must not be placed on a single line. The opening and closing curly brackets must each be placed on their own line.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="AccessModifierMustBeDeclared" ruleCheckId="SA1400">
+        <context>The method must have an access modifier.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Destroy must begin with the 'this.', 'base.', 'object.' or 'TimeManager.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>21</line>
+        <index>529</index>
+        <endIndex>535</endIndex>
+        <startLine>21</startLine>
+        <startColumn>13</startColumn>
+        <endLine>21</endLine>
+        <endColumn>19</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to gameObject must begin with the 'this.', 'base.', 'object.' or 'TimeManager.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>21</line>
+        <index>537</index>
+        <endIndex>546</endIndex>
+        <startLine>21</startLine>
+        <startColumn>21</startColumn>
+        <endLine>21</endLine>
+        <endColumn>30</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="AccessModifierMustBeDeclared" ruleCheckId="SA1400">
+        <context>The method must have an access modifier.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to CurrentTime must begin with the 'this.', 'base.', 'object.' or 'TimeManager.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>26</line>
+        <index>603</index>
+        <endIndex>613</endIndex>
+        <startLine>26</startLine>
+        <startColumn>9</startColumn>
+        <endLine>26</endLine>
+        <endColumn>19</endColumn>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="NavPoint.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.16 20:17:09.487</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="SingleLineCommentMustBePrecededByBlankLine" ruleCheckId="SA1515">
+        <context>A single-line comment must be preceded by a blank line or another single-line comment, or must be the first item in its scope. To ignore this error when commenting out a line of code, begin the comment with '////' rather than '//'.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="SingleLineCommentsMustNotBeFollowedByBlankLine" ruleCheckId="SA1512">
+        <context>A single-line comment must not be followed by a blank line. To ignore this error when commenting out a line of code, begin the comment with '////' rather than '//'.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="SingleLineCommentsMustBeginWithSingleSpace" ruleCheckId="SA1005">
+        <context>The comment must start with a single space. To ignore this error when commenting out a line of code, begin the comment with '////' rather than '//'.</context>
+        <line>14</line>
+        <index>391</index>
+        <endIndex>414</endIndex>
+        <startLine>14</startLine>
+        <startColumn>5</startColumn>
+        <endLine>14</endLine>
+        <endColumn>28</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>property names begin with an upper-case letter: position.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to FlagDistance must begin with the 'this.', 'base.', 'object.' or 'NavPoint.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>18</line>
+        <index>460</index>
+        <endIndex>471</endIndex>
+        <startLine>18</startLine>
+        <startColumn>9</startColumn>
+        <endLine>18</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to position must begin with the 'this.', 'base.', 'object.' or 'NavPoint.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>18</line>
+        <index>538</index>
+        <endIndex>545</endIndex>
+        <startLine>18</startLine>
+        <startColumn>87</startColumn>
+        <endLine>18</endLine>
+        <endColumn>94</endColumn>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="Settings.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.24 01:00:02.216</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="SymbolsMustBeSpacedCorrectly" ruleCheckId="SA1003">
+        <context>The spacing around the symbol '=' is invalid.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>6</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>6</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>9</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>9</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>17</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>17</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>18</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>18</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>19</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>19</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>20</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>20</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>22</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>22</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>25</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>25</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>27</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>27</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>28</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>28</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>29</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>29</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>31</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>31</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>33</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>33</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>35</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>35</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>36</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>36</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>37</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>37</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>38</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.MaintainabilityRules" rule="FieldsMustBePrivate" ruleCheckId="SA1401">
+        <context>Fields must be declared with private access. Use properties to expose fields.</context>
+        <line>38</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="SettingsReader.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 23:29:50.464</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementMustNotBeOnSingleLine" ruleCheckId="SA1502">
+        <context>The property must not be placed on a single line. The opening and closing curly brackets must each be placed on their own line.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All methods must be placed after all properties.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementMustNotBeOnSingleLine" ruleCheckId="SA1502">
+        <context>The property must not be placed on a single line. The opening and closing curly brackets must each be placed on their own line.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to gameSettings must begin with the 'this.', 'base.', 'object.' or 'SettingsReader.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>16</line>
+        <index>419</index>
+        <endIndex>430</endIndex>
+        <startLine>16</startLine>
+        <startColumn>48</startColumn>
+        <endLine>16</endLine>
+        <endColumn>59</endColumn>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="Statistics.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 15:21:15.040</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="OpeningCurlyBracketsMustNotBeFollowedByBlankLine" ruleCheckId="SA1505">
+        <context>An opening curly bracket must not be followed by a blank line.</context>
+        <line>6</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketsMustNotBePrecededByBlankLine" ruleCheckId="SA1508">
+        <context>A closing curly bracket must not be preceded by a blank line.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="AmmoPickUp.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.24 00:59:43.057</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>property names begin with an upper-case letter: type.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>9</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to PickObject must begin with the 'this.', 'base.', 'object.' or 'AmmoPickUp.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>11</line>
+        <index>252</index>
+        <endIndex>261</endIndex>
+        <startLine>11</startLine>
+        <startColumn>9</startColumn>
+        <endLine>11</endLine>
+        <endColumn>18</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="ArmourPickUp.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.24 00:59:39.476</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>property names begin with an upper-case letter: type.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>9</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to PickObject must begin with the 'this.', 'base.', 'object.' or 'ArmourPickUp.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>11</line>
+        <index>250</index>
+        <endIndex>259</endIndex>
+        <startLine>11</startLine>
+        <startColumn>9</startColumn>
+        <endLine>11</endLine>
+        <endColumn>18</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="HealthPickUp.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.24 00:59:35.943</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>property names begin with an upper-case letter: type.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>9</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to PickObject must begin with the 'this.', 'base.', 'object.' or 'HealthPickUp.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>11</line>
+        <index>250</index>
+        <endIndex>259</endIndex>
+        <startLine>11</startLine>
+        <startColumn>9</startColumn>
+        <endLine>11</endLine>
+        <endColumn>18</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="IPickable.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 15:21:15.043</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The interface must have a documentation header.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="ElementMustBeginWithUpperCaseLetter" ruleCheckId="SA1300">
+        <context>property names begin with an upper-case letter: type.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>6</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>6</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="PickUpSpawner.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.24 00:59:47.515</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>7</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>9</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementMustNotBeOnSingleLine" ruleCheckId="SA1502">
+        <context>The property must not be placed on a single line. The opening and closing curly brackets must each be placed on their own line.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>17</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to instance must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>10</line>
+        <index>255</index>
+        <endIndex>262</endIndex>
+        <startLine>10</startLine>
+        <startColumn>50</startColumn>
+        <endLine>10</endLine>
+        <endColumn>57</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>17</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>19</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>22</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the else statement must be wrapped in opening and closing curly brackets.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to instance must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>22</line>
+        <index>612</index>
+        <endIndex>619</endIndex>
+        <startLine>22</startLine>
+        <startColumn>13</startColumn>
+        <endLine>22</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to instance must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>21</line>
+        <index>581</index>
+        <endIndex>588</endIndex>
+        <startLine>21</startLine>
+        <startColumn>13</startColumn>
+        <endLine>21</endLine>
+        <endColumn>20</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to Destroy must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>24</line>
+        <index>656</index>
+        <endIndex>662</endIndex>
+        <startLine>24</startLine>
+        <startColumn>13</startColumn>
+        <endLine>24</endLine>
+        <endColumn>19</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to gameObject must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>24</line>
+        <index>664</index>
+        <endIndex>673</endIndex>
+        <startLine>24</startLine>
+        <startColumn>21</startColumn>
+        <endLine>24</endLine>
+        <endColumn>30</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>27</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the for statement must be wrapped in opening and closing curly brackets.</context>
+        <line>32</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the for statement must be wrapped in opening and closing curly brackets.</context>
+        <line>34</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the foreach statement must be wrapped in opening and closing curly brackets.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>38</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to pickups must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>29</line>
+        <index>728</index>
+        <endIndex>734</endIndex>
+        <startLine>29</startLine>
+        <startColumn>9</startColumn>
+        <endLine>29</endLine>
+        <endColumn>15</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to pickups must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>32</line>
+        <index>890</index>
+        <endIndex>896</endIndex>
+        <startLine>32</startLine>
+        <startColumn>13</startColumn>
+        <endLine>32</endLine>
+        <endColumn>19</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to healthPrefab must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>32</line>
+        <index>925</index>
+        <endIndex>936</endIndex>
+        <startLine>32</startLine>
+        <startColumn>48</startColumn>
+        <endLine>32</endLine>
+        <endColumn>59</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPoints must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>32</line>
+        <index>939</index>
+        <endIndex>949</endIndex>
+        <startLine>32</startLine>
+        <startColumn>62</startColumn>
+        <endLine>32</endLine>
+        <endColumn>72</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPoints must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>32</line>
+        <index>967</index>
+        <endIndex>977</endIndex>
+        <startLine>32</startLine>
+        <startColumn>90</startColumn>
+        <endLine>32</endLine>
+        <endColumn>100</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to pickups must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>34</line>
+        <index>1085</index>
+        <endIndex>1091</endIndex>
+        <startLine>34</startLine>
+        <startColumn>13</startColumn>
+        <endLine>34</endLine>
+        <endColumn>19</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to armourPrefab must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>34</line>
+        <index>1120</index>
+        <endIndex>1131</endIndex>
+        <startLine>34</startLine>
+        <startColumn>48</startColumn>
+        <endLine>34</endLine>
+        <endColumn>59</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPoints must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>34</line>
+        <index>1134</index>
+        <endIndex>1144</endIndex>
+        <startLine>34</startLine>
+        <startColumn>62</startColumn>
+        <endLine>34</endLine>
+        <endColumn>72</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPoints must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>34</line>
+        <index>1162</index>
+        <endIndex>1172</endIndex>
+        <startLine>34</startLine>
+        <startColumn>90</startColumn>
+        <endLine>34</endLine>
+        <endColumn>100</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to pickups must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>37</line>
+        <index>1291</index>
+        <endIndex>1297</endIndex>
+        <startLine>37</startLine>
+        <startColumn>13</startColumn>
+        <endLine>37</endLine>
+        <endColumn>19</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to ammoPrefab must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>37</line>
+        <index>1326</index>
+        <endIndex>1335</endIndex>
+        <startLine>37</startLine>
+        <startColumn>48</startColumn>
+        <endLine>37</endLine>
+        <endColumn>57</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPoints must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>37</line>
+        <index>1338</index>
+        <endIndex>1348</endIndex>
+        <startLine>37</startLine>
+        <startColumn>60</startColumn>
+        <endLine>37</endLine>
+        <endColumn>70</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPoints must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>37</line>
+        <index>1366</index>
+        <endIndex>1376</endIndex>
+        <startLine>37</startLine>
+        <startColumn>88</startColumn>
+        <endLine>37</endLine>
+        <endColumn>98</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to pickups must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>39</line>
+        <index>1473</index>
+        <endIndex>1479</endIndex>
+        <startLine>39</startLine>
+        <startColumn>33</startColumn>
+        <endLine>39</endLine>
+        <endColumn>39</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to StartCoroutine must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>42</line>
+        <index>1531</index>
+        <endIndex>1544</endIndex>
+        <startLine>42</startLine>
+        <startColumn>9</startColumn>
+        <endLine>42</endLine>
+        <endColumn>22</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to SpawnNewPickUps must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>42</line>
+        <index>1546</index>
+        <endIndex>1560</endIndex>
+        <startLine>42</startLine>
+        <startColumn>24</startColumn>
+        <endLine>42</endLine>
+        <endColumn>38</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>45</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>58</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPoints must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>55</line>
+        <index>1898</index>
+        <endIndex>1908</endIndex>
+        <startLine>55</startLine>
+        <startColumn>47</startColumn>
+        <endLine>55</endLine>
+        <endColumn>57</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to spawnPoints must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>55</line>
+        <index>1926</index>
+        <endIndex>1936</endIndex>
+        <startLine>55</startLine>
+        <startColumn>75</startColumn>
+        <endLine>55</endLine>
+        <endColumn>85</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to IsDisableCheck must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>50</line>
+        <index>1703</index>
+        <endIndex>1716</endIndex>
+        <startLine>50</startLine>
+        <startColumn>16</startColumn>
+        <endLine>50</endLine>
+        <endColumn>29</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="KeywordsMustBeSpacedCorrectly" ruleCheckId="SA1000">
+        <context>The spacing around the keyword 'while' is invalid.</context>
+        <line>47</line>
+        <index>1634</index>
+        <endIndex>1638</endIndex>
+        <startLine>47</startLine>
+        <startColumn>9</startColumn>
+        <endLine>47</endLine>
+        <endColumn>13</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="KeywordsMustBeSpacedCorrectly" ruleCheckId="SA1000">
+        <context>The spacing around the keyword 'if' is invalid.</context>
+        <line>50</line>
+        <index>1700</index>
+        <endIndex>1701</endIndex>
+        <startLine>50</startLine>
+        <startColumn>13</startColumn>
+        <endLine>50</endLine>
+        <endColumn>14</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="CodeMustNotContainMultipleWhitespaceInARow" ruleCheckId="SA1025">
+        <context>The code contains multiple spaces in a row. Only one space is needed.</context>
+        <line>59</line>
+        <index>2062</index>
+        <endIndex>2063</endIndex>
+        <startLine>59</startLine>
+        <startColumn>29</startColumn>
+        <endLine>59</endLine>
+        <endColumn>30</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>63</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>72</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to pickups must begin with the 'this.', 'base.', 'object.' or 'PickUpSpawner.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>65</line>
+        <index>2196</index>
+        <endIndex>2202</endIndex>
+        <startLine>65</startLine>
+        <startColumn>29</startColumn>
+        <endLine>65</endLine>
+        <endColumn>35</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="KeywordsMustBeSpacedCorrectly" ruleCheckId="SA1000">
+        <context>The spacing around the keyword 'foreach' is invalid.</context>
+        <line>65</line>
+        <index>2176</index>
+        <endIndex>2182</endIndex>
+        <startLine>65</startLine>
+        <startColumn>9</startColumn>
+        <endLine>65</endLine>
+        <endColumn>15</endColumn>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="PickUpType.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 15:21:15.044</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The enum must have a documentation header.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="Sensors.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 22:38:30.693</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CodeMustNotContainBlankLinesAtEndOfFile" ruleCheckId="SA1518">
+        <context>The code must not contain blank lines at the end of the file.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CodeMustNotContainMultipleBlankLinesInARow" ruleCheckId="SA1507">
+        <context>The code must not contain multiple blank lines in a row.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="SensorType.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 15:21:15.046</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The enum must have a documentation header.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="EnumerationItemsMustBeDocumented" ruleCheckId="SA1602">
+        <context>The enumeration sub-item must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="BoolToInteger.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.24 00:44:55.312</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.NamingRules" rule="FieldNamesMustNotBeginWithUnderscore" ruleCheckId="SA1309">
+        <context>Field names must not start with an underscore.</context>
+        <line>3</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="SerializableDictionary.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 16:12:15.866</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>15</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>17</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>18</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>19</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>20</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>21</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>23</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>26</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>31</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The indexer must have a documentation header.</context>
+        <line>36</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The indexer must have a documentation header.</context>
+        <line>47</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The constructor must have a documentation header.</context>
+        <line>60</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The constructor must have a documentation header.</context>
+        <line>65</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The constructor must have a documentation header.</context>
+        <line>70</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The constructor must have a documentation header.</context>
+        <line>75</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The constructor must have a documentation header.</context>
+        <line>85</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The constructor must have a documentation header.</context>
+        <line>90</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>100</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>122</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>127</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>146</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>151</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>190</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>195</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>226</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>281</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>297</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>314</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>325</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>327</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>403</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>420</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>439</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>450</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>455</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>460</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>465</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>472</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>490</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>495</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>500</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>505</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>510</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <project key="-364018800">
+    <configuration>UNITY_2019_4_35;UNITY_2019_4;UNITY_2019;UNITY_5_3_OR_NEWER;UNITY_5_4_OR_NEWER;UNITY_5_5_OR_NEWER;UNITY_5_6_OR_NEWER;UNITY_2017_1_OR_NEWER;UNITY_2017_2_OR_NEWER;UNITY_2017_3_OR_NEWER;UNITY_2017_4_OR_NEWER;UNITY_2018_1_OR_NEWER;UNITY_2018_2_OR_NEWER;UNITY_2018_3_OR_NEWER;UNITY_2018_4_OR_NEWER;UNITY_2019_1_OR_NEWER;UNITY_2019_2_OR_NEWER;UNITY_2019_3_OR_NEWER;UNITY_2019_4_OR_NEWER;PLATFORM_ARCH_64;UNITY_64;UNITY_INCLUDE_TESTS;ENABLE_AR;ENABLE_AUDIO;ENABLE_CACHING;ENABLE_CLOTH;ENABLE_EVENT_QUEUE;ENABLE_MICROPHONE;ENABLE_MULTIPLE_DISPLAYS;ENABLE_PHYSICS;ENABLE_TEXTURE_STREAMING;ENABLE_UNET;ENABLE_LZMA;ENABLE_UNITYEVENTS;ENABLE_VR;ENABLE_WEBCAM;ENABLE_UNITYWEBREQUEST;ENABLE_WWW;ENABLE_CLOUD_SERVICES;ENABLE_CLOUD_SERVICES_COLLAB;ENABLE_CLOUD_SERVICES_COLLAB_SOFTLOCKS;ENABLE_CLOUD_SERVICES_ADS;ENABLE_CLOUD_SERVICES_USE_WEBREQUEST;ENABLE_CLOUD_SERVICES_CRASH_REPORTING;ENABLE_CLOUD_SERVICES_PURCHASING;ENABLE_CLOUD_SERVICES_ANALYTICS;ENABLE_CLOUD_SERVICES_UNET;ENABLE_CLOUD_SERVICES_BUILD;ENABLE_CLOUD_LICENSE;ENABLE_EDITOR_HUB_LICENSE;ENABLE_WEBSOCKET_CLIENT;ENABLE_DIRECTOR_AUDIO;ENABLE_DIRECTOR_TEXTURE;ENABLE_MANAGED_JOBS;ENABLE_MANAGED_TRANSFORM_JOBS;ENABLE_MANAGED_ANIMATION_JOBS;ENABLE_MANAGED_AUDIO_JOBS;INCLUDE_DYNAMIC_GI;ENABLE_MONO_BDWGC;ENABLE_SCRIPTING_GC_WBARRIERS;PLATFORM_SUPPORTS_MONO;RENDER_SOFTWARE_CURSOR;ENABLE_VIDEO;PLATFORM_STANDALONE;PLATFORM_STANDALONE_WIN;UNITY_STANDALONE_WIN;UNITY_STANDALONE;ENABLE_RUNTIME_GI;ENABLE_MOVIES;ENABLE_NETWORK;ENABLE_CRUNCH_TEXTURE_COMPRESSION;ENABLE_OUT_OF_PROCESS_CRASH_HANDLER;ENABLE_CLUSTER_SYNC;ENABLE_CLUSTERINPUT;GFXDEVICE_WAITFOREVENT_MESSAGEPUMP;ENABLE_WEBSOCKET_HOST;ENABLE_MONO;NET_4_6;ENABLE_PROFILER;DEBUG;TRACE;UNITY_ASSERTIONS;UNITY_EDITOR;UNITY_EDITOR_64;UNITY_EDITOR_WIN;ENABLE_UNITY_COLLECTIONS_CHECKS;ENABLE_BURST_AOT;UNITY_TEAM_LICENSE;ENABLE_CUSTOM_RENDER_TEXTURE;ENABLE_DIRECTOR;ENABLE_LOCALIZATION;ENABLE_SPRITES;ENABLE_TERRAIN;ENABLE_TILEMAP;ENABLE_TIMELINE;ENABLE_INPUT_SYSTEM;ENABLE_LEGACY_INPUT_MANAGER;CSHARP_7_OR_LATER;CSHARP_7_3_OR_NEWER</configuration>
+  </project>
+  <sourcecode name="NPC.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.28 18:30:19.041</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.CsParser" rule="SyntaxException" ruleCheckId="SA0102">
+        <context>A syntax error has been discovered in file F:\SigmaRiskManagment\real shooter Git Version\Assets\Scripts\Character\NPC.cs on line 87.</context>
+        <line>87</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="FlagZone.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.28 18:30:19.035</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>5</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>8</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.OrderingRules" rule="ElementsMustAppearInTheCorrectOrder" ruleCheckId="SA1201">
+        <context>All properties must be placed after all fields.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>9</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>9</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>10</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>11</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to occupDefenders must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>12</line>
+        <index>362</index>
+        <endIndex>375</endIndex>
+        <startLine>12</startLine>
+        <startColumn>33</startColumn>
+        <endLine>12</endLine>
+        <endColumn>46</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to occupAttackers must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>12</line>
+        <index>384</index>
+        <endIndex>397</endIndex>
+        <startLine>12</startLine>
+        <startColumn>55</startColumn>
+        <endLine>12</endLine>
+        <endColumn>68</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="SymbolsMustBeSpacedCorrectly" ruleCheckId="SA1003">
+        <context>The spacing around the symbol '&gt;' is invalid.</context>
+        <line>12</line>
+        <index>376</index>
+        <endIndex>376</endIndex>
+        <startLine>12</startLine>
+        <startColumn>47</startColumn>
+        <endLine>12</endLine>
+        <endColumn>47</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="SymbolsMustBeSpacedCorrectly" ruleCheckId="SA1003">
+        <context>The spacing around the symbol '&gt;' is invalid.</context>
+        <line>12</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="SymbolsMustBeSpacedCorrectly" ruleCheckId="SA1003">
+        <context>The spacing around the symbol '&gt;' is invalid.</context>
+        <line>12</line>
+        <index>398</index>
+        <endIndex>398</endIndex>
+        <startLine>12</startLine>
+        <startColumn>69</startColumn>
+        <endLine>12</endLine>
+        <endColumn>69</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The property must have a documentation header.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>13</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to occupDefenders must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>13</line>
+        <index>435</index>
+        <endIndex>448</endIndex>
+        <startLine>13</startLine>
+        <startColumn>32</startColumn>
+        <endLine>13</endLine>
+        <endColumn>45</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to occupAttackers must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>13</line>
+        <index>460</index>
+        <endIndex>473</endIndex>
+        <startLine>13</startLine>
+        <startColumn>57</startColumn>
+        <endLine>13</endLine>
+        <endColumn>70</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The field must have a documentation header.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>14</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>16</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="OpeningCurlyBracketsMustNotBeFollowedByBlankLine" ruleCheckId="SA1505">
+        <context>An opening curly bracket must not be followed by a blank line.</context>
+        <line>17</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>24</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to timeForWin must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>19</line>
+        <index>559</index>
+        <endIndex>568</endIndex>
+        <startLine>19</startLine>
+        <startColumn>9</startColumn>
+        <endLine>19</endLine>
+        <endColumn>18</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to TimeStayAttackers must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>20</line>
+        <index>628</index>
+        <endIndex>644</endIndex>
+        <startLine>20</startLine>
+        <startColumn>9</startColumn>
+        <endLine>20</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to TimeStayDefenders must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>21</line>
+        <index>660</index>
+        <endIndex>676</endIndex>
+        <startLine>21</startLine>
+        <startColumn>9</startColumn>
+        <endLine>21</endLine>
+        <endColumn>25</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to occupAttackers must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>22</line>
+        <index>692</index>
+        <endIndex>705</endIndex>
+        <startLine>22</startLine>
+        <startColumn>9</startColumn>
+        <endLine>22</endLine>
+        <endColumn>22</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to occupDefenders must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>23</line>
+        <index>721</index>
+        <endIndex>734</endIndex>
+        <startLine>23</startLine>
+        <startColumn>9</startColumn>
+        <endLine>23</endLine>
+        <endColumn>22</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>25</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>25</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>39</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to occupDefenders must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>30</line>
+        <index>889</index>
+        <endIndex>902</endIndex>
+        <startLine>30</startLine>
+        <startColumn>17</startColumn>
+        <endLine>30</endLine>
+        <endColumn>30</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to occupAttackers must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>33</line>
+        <index>978</index>
+        <endIndex>991</endIndex>
+        <startLine>33</startLine>
+        <startColumn>17</startColumn>
+        <endLine>33</endLine>
+        <endColumn>30</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.SpacingRules" rule="KeywordsMustBeSpacedCorrectly" ruleCheckId="SA1000">
+        <context>The spacing around the keyword 'switch' is invalid.</context>
+        <line>27</line>
+        <index>813</index>
+        <endIndex>818</endIndex>
+        <startLine>27</startLine>
+        <startColumn>9</startColumn>
+        <endLine>27</endLine>
+        <endColumn>14</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>40</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ClosingCurlyBracketMustBeFollowedByBlankLine" ruleCheckId="SA1513">
+        <context>Statements or elements wrapped in curly brackets must be followed by a blank line.</context>
+        <line>54</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to occupDefenders must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>45</line>
+        <index>1287</index>
+        <endIndex>1300</endIndex>
+        <startLine>45</startLine>
+        <startColumn>17</startColumn>
+        <endLine>45</endLine>
+        <endColumn>30</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to occupAttackers must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>48</line>
+        <index>1376</index>
+        <endIndex>1389</endIndex>
+        <startLine>48</startLine>
+        <startColumn>17</startColumn>
+        <endLine>48</endLine>
+        <endColumn>30</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The method must have a documentation header.</context>
+        <line>55</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="ElementsMustBeSeparatedByBlankLine" ruleCheckId="SA1516">
+        <context>Adjacent elements must be separated by a blank line.</context>
+        <line>55</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>67</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.LayoutRules" rule="CurlyBracketsMustNotBeOmitted" ruleCheckId="SA1503">
+        <context>The body of the if statement must be wrapped in opening and closing curly brackets.</context>
+        <line>73</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to TimeStayAttackers must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>59</line>
+        <index>1641</index>
+        <endIndex>1657</endIndex>
+        <startLine>59</startLine>
+        <startColumn>13</startColumn>
+        <endLine>59</endLine>
+        <endColumn>29</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to TimeStayDefenders must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>60</line>
+        <index>1677</index>
+        <endIndex>1693</endIndex>
+        <startLine>60</startLine>
+        <startColumn>13</startColumn>
+        <endLine>60</endLine>
+        <endColumn>29</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to IsOccupBoth must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>57</line>
+        <index>1590</index>
+        <endIndex>1600</endIndex>
+        <startLine>57</startLine>
+        <startColumn>13</startColumn>
+        <endLine>57</endLine>
+        <endColumn>23</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to IsNotOccup must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>57</line>
+        <index>1605</index>
+        <endIndex>1614</endIndex>
+        <startLine>57</startLine>
+        <startColumn>28</startColumn>
+        <endLine>57</endLine>
+        <endColumn>37</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to TimeStayAttackers must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>65</line>
+        <index>1794</index>
+        <endIndex>1810</endIndex>
+        <startLine>65</startLine>
+        <startColumn>13</startColumn>
+        <endLine>65</endLine>
+        <endColumn>29</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to TimeStayAttackers must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>66</line>
+        <index>1848</index>
+        <endIndex>1864</endIndex>
+        <startLine>66</startLine>
+        <startColumn>17</startColumn>
+        <endLine>66</endLine>
+        <endColumn>33</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to timeForWin must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>66</line>
+        <index>1868</index>
+        <endIndex>1877</endIndex>
+        <startLine>66</startLine>
+        <startColumn>37</startColumn>
+        <endLine>66</endLine>
+        <endColumn>46</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to occupAttackers must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>63</line>
+        <index>1750</index>
+        <endIndex>1763</endIndex>
+        <startLine>63</startLine>
+        <startColumn>18</startColumn>
+        <endLine>63</endLine>
+        <endColumn>31</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to TimeStayDefenders must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>71</line>
+        <index>1999</index>
+        <endIndex>2015</endIndex>
+        <startLine>71</startLine>
+        <startColumn>13</startColumn>
+        <endLine>71</endLine>
+        <endColumn>29</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to TimeStayDefenders must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>72</line>
+        <index>2053</index>
+        <endIndex>2069</endIndex>
+        <startLine>72</startLine>
+        <startColumn>17</startColumn>
+        <endLine>72</endLine>
+        <endColumn>33</endColumn>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.ReadabilityRules" rule="PrefixCallsCorrectly" ruleCheckId="SA1126">
+        <context>The call to timeForWin must begin with the 'this.', 'base.', 'object.' or 'FlagZone.' or 'MonoBehaviour.' prefix to indicate the intended method call. </context>
+        <line>72</line>
+        <index>2073</index>
+        <endIndex>2082</endIndex>
+        <startLine>72</startLine>
+        <startColumn>37</startColumn>
+        <endLine>72</endLine>
+        <endColumn>46</endColumn>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+  <sourcecode name="scr_WeaponController.cs" parser="StyleCop.CSharp.CsParser">
+    <timestamps>
+      <styleCop>2022.04.28 18:27:01.731</styleCop>
+      <settingsFile>2022.04.28 18:27:01.710</settingsFile>
+      <sourceFile>2022.04.12 15:21:15.050</sourceFile>
+      <parser>2022.04.28 18:27:01.731</parser>
+      <StyleCop.CSharp.DocumentationRules>2022.04.28 18:27:01.731</StyleCop.CSharp.DocumentationRules>
+      <StyleCop.CSharp.DocumentationRules.FilesHashCode>-2051395988</StyleCop.CSharp.DocumentationRules.FilesHashCode>
+      <StyleCop.CSharp.LayoutRules>2022.04.28 18:27:01.731</StyleCop.CSharp.LayoutRules>
+      <StyleCop.CSharp.LayoutRules.FilesHashCode>0</StyleCop.CSharp.LayoutRules.FilesHashCode>
+      <StyleCop.CSharp.MaintainabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.MaintainabilityRules>
+      <StyleCop.CSharp.MaintainabilityRules.FilesHashCode>0</StyleCop.CSharp.MaintainabilityRules.FilesHashCode>
+      <StyleCop.CSharp.NamingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.NamingRules>
+      <StyleCop.CSharp.NamingRules.FilesHashCode>0</StyleCop.CSharp.NamingRules.FilesHashCode>
+      <StyleCop.CSharp.OrderingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.OrderingRules>
+      <StyleCop.CSharp.OrderingRules.FilesHashCode>0</StyleCop.CSharp.OrderingRules.FilesHashCode>
+      <StyleCop.CSharp.ReadabilityRules>2022.04.28 18:27:01.731</StyleCop.CSharp.ReadabilityRules>
+      <StyleCop.CSharp.ReadabilityRules.FilesHashCode>0</StyleCop.CSharp.ReadabilityRules.FilesHashCode>
+      <StyleCop.CSharp.SpacingRules>2022.04.28 18:27:01.731</StyleCop.CSharp.SpacingRules>
+      <StyleCop.CSharp.SpacingRules.FilesHashCode>0</StyleCop.CSharp.SpacingRules.FilesHashCode>
+    </timestamps>
+    <violations>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="FileMustHaveHeader" ruleCheckId="SA1633">
+        <context>The file has no header, the header Xml is invalid, or the header is not located at the top of the file.</context>
+        <line>1</line>
+        <warning>False</warning>
+      </violation>
+      <violation namespace="StyleCop.CSharp.DocumentationRules" rule="ElementsMustBeDocumented" ruleCheckId="SA1600">
+        <context>The class must have a documentation header.</context>
+        <line>4</line>
+        <warning>False</warning>
+      </violation>
+    </violations>
+  </sourcecode>
+</stylecopresultscache>
\ No newline at end of file
-- 
2.49.0


From 690f1e253602c6fded329ceca14cc25d3b4ce70f Mon Sep 17 00:00:00 2001
From: Andrey Gumirov <gavt45@bk.ru>
Date: Thu, 5 May 2022 01:08:58 +0700
Subject: [PATCH 05/17] test

---
 1 | 0
 1 file changed, 0 insertions(+), 0 deletions(-)
 create mode 100644 1

diff --git a/1 b/1
new file mode 100644
index 0000000..e69de29
-- 
2.49.0


From 351fa8ee12c499a0374be71a192a5f288611f467 Mon Sep 17 00:00:00 2001
From: Andrey Gumirov <gavt45@bk.ru>
Date: Thu, 5 May 2022 01:10:04 +0700
Subject: [PATCH 06/17] test 1

---
 1 | 1 +
 1 file changed, 1 insertion(+)

diff --git a/1 b/1
index e69de29..8b13789 100644
--- a/1
+++ b/1
@@ -0,0 +1 @@
+
-- 
2.49.0


From cfb7b71a8a61ae44419070002cfd73049c278d93 Mon Sep 17 00:00:00 2001
From: Andrey Gumirov <gavt45@bk.ru>
Date: Thu, 5 May 2022 15:37:51 +0700
Subject: [PATCH 07/17] Fixed merge errors

---
 .../Scripts/Character/CharacterCondition.cs   | 10 -----
 .../Scripts/Character/MovementController.cs   | 25 -----------
 Assets/Scripts/Character/NPC.cs               | 45 -------------------
 Assets/Scripts/Misc/NavPoint.cs               |  9 ----
 4 files changed, 89 deletions(-)

diff --git a/Assets/Scripts/Character/CharacterCondition.cs b/Assets/Scripts/Character/CharacterCondition.cs
index 1efa341..de60740 100755
--- a/Assets/Scripts/Character/CharacterCondition.cs
+++ b/Assets/Scripts/Character/CharacterCondition.cs
@@ -9,22 +9,12 @@ public class CharacterCondition
     public event Action<int> OnChangeAmmunitionEvent;
 
     private int health;
-<<<<<<< HEAD
     public int HealthPoints
     {
         get
         {
             return health;
         }
-=======
-
-    public int HealthPoints 
-    { 
-        get 
-        { 
-            return health; 
-        } 
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
         private set
         {
             health = value;
diff --git a/Assets/Scripts/Character/MovementController.cs b/Assets/Scripts/Character/MovementController.cs
index 3ade88a..0022357 100644
--- a/Assets/Scripts/Character/MovementController.cs
+++ b/Assets/Scripts/Character/MovementController.cs
@@ -6,18 +6,11 @@ using UnityEngine.AI;
 [RequireComponent(typeof(NavMeshAgent))]
 public class MovementController : MonoBehaviour
 {
-<<<<<<< HEAD
     public int PointStartID { get; set; }
     public int PointEndID { get; private set; }
     public float FlagDistance { get; private set; }
     private const float updateFlagPositionDelay = 5;
     private const float updateReachedDestinationDelay = 5;
-=======
-    public NavPoint currentPosition { get; set; }
-    private Dictionary<int, NavPoint> navPoints = new Dictionary<int, NavPoint>();
-    
-    [SerializeField] private NavMeshAgent navMeshAgent;
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
 
     [SerializeField] private NavMeshAgent navMeshAgent;
     [SerializeField] private GameObject flag;
@@ -28,7 +21,6 @@ public class MovementController : MonoBehaviour
 
     private void Awake()
     {
-<<<<<<< HEAD
         navMeshAgent.speed = SettingsReader.Instance.GetSettings.MovementSpeed;
         idNavPointDict = MapManager.IDToNavPoint;
         InvokeRepeating(nameof(UpdateFlagPosition), 0, updateFlagPositionDelay);
@@ -43,12 +35,6 @@ public class MovementController : MonoBehaviour
     private void UpdateFlagPosition()
     {
         FlagDistance = (flag.transform.position - gameObject.transform.position).magnitude;
-=======
-        navMeshAgent.speed = SettingsReader.Instance.GetSettings.movementSpeed;
-        foreach (var np in MapManager.navPoints) {
-            navPoints[np.PointId] = np;
-        }
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
     }
 
     public void MoveToPointById(int id)
@@ -62,27 +48,16 @@ public class MovementController : MonoBehaviour
     }
     
     public void MoveToRandomPoint()
-<<<<<<< HEAD
     {
         Debug.Log(MapManager.NavPoints == null);
         GoToNextNavPoint(MapManager.NavPoints[Random.Range(0, MapManager.NavPoints.Count)]);
-=======
-    {   
-        // Debug.Log(MapManager.navPoints == null);
-        goToNextNavPoint(MapManager.navPoints[Random.Range(0, MapManager.navPoints.Count)]);
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
     }
 
     public List<NavPoint> GetPointsCandidate()
     {
-<<<<<<< HEAD
         return MapManager.NavPoints
             .Where(point => 
             (idNavPointDict[PointStartID].Position - point.Position).magnitude < SettingsReader.Instance.GetSettings.MovementDistance)
-=======
-        return MapManager.navPoints
-            .Where(point => (currentPosition.position - point.position).magnitude < SettingsReader.Instance.GetSettings.movementDistance)
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
             .ToList();
     }
 
diff --git a/Assets/Scripts/Character/NPC.cs b/Assets/Scripts/Character/NPC.cs
index 932379c..d47e377 100644
--- a/Assets/Scripts/Character/NPC.cs
+++ b/Assets/Scripts/Character/NPC.cs
@@ -70,14 +70,7 @@ public class NPC : Agent, ICharacter
 
     public override void CollectObservations(VectorSensor sensor)
     {
-<<<<<<< HEAD
         var candidates = moveController.GetPointsCandidate();
-=======
-        sensor.AddObservation(Condition.HealthPoints);
-        sensor.AddObservation(Condition.ArmourPoints);
-        sensor.AddObservation(Condition.Ammunition);
-        sensor.AddObservation((int) NPC_State.State);
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
 
         //common sensors
         sensor.AddObservation(GameManager.IsHaveSeenByEnemy(AgentCharacter.Team.GetOppositeTeam(),
@@ -100,7 +93,6 @@ public class NPC : Agent, ICharacter
         //point sensors
         foreach (var point in candidates)
         {
-<<<<<<< HEAD
             bufferSensor.AppendObservation(new float[] {
                 point.DeathAttr,
                 (int)point.navType,
@@ -112,35 +104,6 @@ public class NPC : Agent, ICharacter
                 GameManager.IsHaveSeenByEnemy(AgentCharacter.Team.GetOppositeTeam(),
                               point.Position).ToInt()
         });
-=======
-            var parray = new float[]
-            {
-                //1 position in navpointId
-                (float) moveController.currentPosition.PointId,
-                //2 distance to flag
-                moveController.currentPosition.FlagDistance,
-                //3 death count in point
-                moveController.currentPosition.DeathAttr,
-                //4 flagEnemyDistance
-                GameManager.IsCloserToFlagFromNextNavPoint(point, transform.position) == true ? 1 : 0,
-                //5 EnemyVsNavPointDistance
-                GameManager.IsCloserToEnemyThanToNextNavPoint(point, transform.position, AgentCharacter.Team) == true
-                    ? 1
-                    : 0
-            };
-            // var _parray = string.Join(" ", parray);
-            // Debug.Log("OBS: " + _parray);
-            bufferSensor.AppendObservation(parray);
-        }
-    }
-
-    public override void Heuristic(in ActionBuffers actionsOut)
-    {
-        var discreteActionsOut = actionsOut.DiscreteActions;
-        if (Input.GetKeyDown(KeyCode.W))
-        {
-            discreteActionsOut[0] = 1;
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
         }
     }
 
@@ -149,7 +112,6 @@ public class NPC : Agent, ICharacter
         var result = actions.DiscreteActions;
         if (result[0] == 0)
         {
-<<<<<<< HEAD
             if (navPointIdDict[moveController.PointStartID].navType != NavPointType.Cover)
                 return;
             NpcState = CoverState;
@@ -185,13 +147,6 @@ public class NPC : Agent, ICharacter
                 case 1: moveController.ReturnToStartPoint(); NpcState = RunningState; break;
                 default: throw new ArgumentException("Undefined Action recieved");
             }
-=======
-            moveController.MoveToRandomPoint();
-            NPC_State = RunningState;
-        } else if (actions.DiscreteActions[0] == 2)
-        {
-            moveController.MoveToPointById(actions.DiscreteActions[1]);
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
         }
     }
     #endregion
diff --git a/Assets/Scripts/Misc/NavPoint.cs b/Assets/Scripts/Misc/NavPoint.cs
index 473d7fa..a70fd61 100755
--- a/Assets/Scripts/Misc/NavPoint.cs
+++ b/Assets/Scripts/Misc/NavPoint.cs
@@ -17,21 +17,12 @@ public class NavPoint : MonoBehaviour
     public NavPointType navType = NavPointType.Direction;
 
     [HideInInspector]
-<<<<<<< HEAD
     public int PointId = 0;
-=======
-    public int PointId;
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
     public float DeathAttr = 0;
     public List<Vector3> EnemiesSeen = new List<Vector3>();
 
     private void Start()
     {
-<<<<<<< HEAD
         FlagDistance = (GameObject.FindGameObjectWithTag("Flag").transform.position - Position).magnitude;
-=======
-        PointId = GetInstanceID();
-        FlagDistance = (GameObject.FindGameObjectWithTag("Flag").transform.position - position).magnitude;               
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
     }
 }
-- 
2.49.0


From f3fe5ea4cd8ef270fbfc6c1be90b25b6576f5a93 Mon Sep 17 00:00:00 2001
From: Krazerleo <krazerleo@gmail.com>
Date: Thu, 5 May 2022 16:26:23 +0700
Subject: [PATCH 08/17] commit 665

---
 Assets/Prefabs/Player.prefab.meta | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Assets/Prefabs/Player.prefab.meta b/Assets/Prefabs/Player.prefab.meta
index 9180d60..1fa6659 100755
--- a/Assets/Prefabs/Player.prefab.meta
+++ b/Assets/Prefabs/Player.prefab.meta
@@ -1,5 +1,5 @@
 fileFormatVersion: 2
-guid: 99bb17a23a489624baeaf337f91a4f84
+guid: 80f6c1c85e5daed4c96c70205ed5503d
 PrefabImporter:
   externalObjects: {}
   userData: 
-- 
2.49.0


From ba1b350c6283b0ebb3d6e48bf7de66db8f6503a6 Mon Sep 17 00:00:00 2001
From: Krazerleo <krazerleo@gmail.com>
Date: Thu, 5 May 2022 16:49:21 +0700
Subject: [PATCH 09/17] resolved merge conflicts

---
 .../Scripts/Character/CharacterCondition.cs   |  9 ----
 .../Scripts/Character/MovementController.cs   | 26 -----------
 Assets/Scripts/Character/NPC.cs               | 45 -------------------
 Assets/Scripts/Misc/NavPoint.cs               |  9 ----
 4 files changed, 89 deletions(-)

diff --git a/Assets/Scripts/Character/CharacterCondition.cs b/Assets/Scripts/Character/CharacterCondition.cs
index 1efa341..8d7d5c5 100755
--- a/Assets/Scripts/Character/CharacterCondition.cs
+++ b/Assets/Scripts/Character/CharacterCondition.cs
@@ -9,14 +9,6 @@ public class CharacterCondition
     public event Action<int> OnChangeAmmunitionEvent;
 
     private int health;
-<<<<<<< HEAD
-    public int HealthPoints
-    {
-        get
-        {
-            return health;
-        }
-=======
 
     public int HealthPoints 
     { 
@@ -24,7 +16,6 @@ public class CharacterCondition
         { 
             return health; 
         } 
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
         private set
         {
             health = value;
diff --git a/Assets/Scripts/Character/MovementController.cs b/Assets/Scripts/Character/MovementController.cs
index 3ade88a..b1849ce 100644
--- a/Assets/Scripts/Character/MovementController.cs
+++ b/Assets/Scripts/Character/MovementController.cs
@@ -6,18 +6,11 @@ using UnityEngine.AI;
 [RequireComponent(typeof(NavMeshAgent))]
 public class MovementController : MonoBehaviour
 {
-<<<<<<< HEAD
     public int PointStartID { get; set; }
     public int PointEndID { get; private set; }
     public float FlagDistance { get; private set; }
     private const float updateFlagPositionDelay = 5;
     private const float updateReachedDestinationDelay = 5;
-=======
-    public NavPoint currentPosition { get; set; }
-    private Dictionary<int, NavPoint> navPoints = new Dictionary<int, NavPoint>();
-    
-    [SerializeField] private NavMeshAgent navMeshAgent;
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
 
     [SerializeField] private NavMeshAgent navMeshAgent;
     [SerializeField] private GameObject flag;
@@ -28,7 +21,6 @@ public class MovementController : MonoBehaviour
 
     private void Awake()
     {
-<<<<<<< HEAD
         navMeshAgent.speed = SettingsReader.Instance.GetSettings.MovementSpeed;
         idNavPointDict = MapManager.IDToNavPoint;
         InvokeRepeating(nameof(UpdateFlagPosition), 0, updateFlagPositionDelay);
@@ -43,12 +35,6 @@ public class MovementController : MonoBehaviour
     private void UpdateFlagPosition()
     {
         FlagDistance = (flag.transform.position - gameObject.transform.position).magnitude;
-=======
-        navMeshAgent.speed = SettingsReader.Instance.GetSettings.movementSpeed;
-        foreach (var np in MapManager.navPoints) {
-            navPoints[np.PointId] = np;
-        }
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
     }
 
     public void MoveToPointById(int id)
@@ -62,27 +48,15 @@ public class MovementController : MonoBehaviour
     }
     
     public void MoveToRandomPoint()
-<<<<<<< HEAD
     {
-        Debug.Log(MapManager.NavPoints == null);
         GoToNextNavPoint(MapManager.NavPoints[Random.Range(0, MapManager.NavPoints.Count)]);
-=======
-    {   
-        // Debug.Log(MapManager.navPoints == null);
-        goToNextNavPoint(MapManager.navPoints[Random.Range(0, MapManager.navPoints.Count)]);
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
     }
 
     public List<NavPoint> GetPointsCandidate()
     {
-<<<<<<< HEAD
         return MapManager.NavPoints
             .Where(point => 
             (idNavPointDict[PointStartID].Position - point.Position).magnitude < SettingsReader.Instance.GetSettings.MovementDistance)
-=======
-        return MapManager.navPoints
-            .Where(point => (currentPosition.position - point.position).magnitude < SettingsReader.Instance.GetSettings.movementDistance)
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
             .ToList();
     }
 
diff --git a/Assets/Scripts/Character/NPC.cs b/Assets/Scripts/Character/NPC.cs
index 932379c..d47e377 100644
--- a/Assets/Scripts/Character/NPC.cs
+++ b/Assets/Scripts/Character/NPC.cs
@@ -70,14 +70,7 @@ public class NPC : Agent, ICharacter
 
     public override void CollectObservations(VectorSensor sensor)
     {
-<<<<<<< HEAD
         var candidates = moveController.GetPointsCandidate();
-=======
-        sensor.AddObservation(Condition.HealthPoints);
-        sensor.AddObservation(Condition.ArmourPoints);
-        sensor.AddObservation(Condition.Ammunition);
-        sensor.AddObservation((int) NPC_State.State);
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
 
         //common sensors
         sensor.AddObservation(GameManager.IsHaveSeenByEnemy(AgentCharacter.Team.GetOppositeTeam(),
@@ -100,7 +93,6 @@ public class NPC : Agent, ICharacter
         //point sensors
         foreach (var point in candidates)
         {
-<<<<<<< HEAD
             bufferSensor.AppendObservation(new float[] {
                 point.DeathAttr,
                 (int)point.navType,
@@ -112,35 +104,6 @@ public class NPC : Agent, ICharacter
                 GameManager.IsHaveSeenByEnemy(AgentCharacter.Team.GetOppositeTeam(),
                               point.Position).ToInt()
         });
-=======
-            var parray = new float[]
-            {
-                //1 position in navpointId
-                (float) moveController.currentPosition.PointId,
-                //2 distance to flag
-                moveController.currentPosition.FlagDistance,
-                //3 death count in point
-                moveController.currentPosition.DeathAttr,
-                //4 flagEnemyDistance
-                GameManager.IsCloserToFlagFromNextNavPoint(point, transform.position) == true ? 1 : 0,
-                //5 EnemyVsNavPointDistance
-                GameManager.IsCloserToEnemyThanToNextNavPoint(point, transform.position, AgentCharacter.Team) == true
-                    ? 1
-                    : 0
-            };
-            // var _parray = string.Join(" ", parray);
-            // Debug.Log("OBS: " + _parray);
-            bufferSensor.AppendObservation(parray);
-        }
-    }
-
-    public override void Heuristic(in ActionBuffers actionsOut)
-    {
-        var discreteActionsOut = actionsOut.DiscreteActions;
-        if (Input.GetKeyDown(KeyCode.W))
-        {
-            discreteActionsOut[0] = 1;
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
         }
     }
 
@@ -149,7 +112,6 @@ public class NPC : Agent, ICharacter
         var result = actions.DiscreteActions;
         if (result[0] == 0)
         {
-<<<<<<< HEAD
             if (navPointIdDict[moveController.PointStartID].navType != NavPointType.Cover)
                 return;
             NpcState = CoverState;
@@ -185,13 +147,6 @@ public class NPC : Agent, ICharacter
                 case 1: moveController.ReturnToStartPoint(); NpcState = RunningState; break;
                 default: throw new ArgumentException("Undefined Action recieved");
             }
-=======
-            moveController.MoveToRandomPoint();
-            NPC_State = RunningState;
-        } else if (actions.DiscreteActions[0] == 2)
-        {
-            moveController.MoveToPointById(actions.DiscreteActions[1]);
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
         }
     }
     #endregion
diff --git a/Assets/Scripts/Misc/NavPoint.cs b/Assets/Scripts/Misc/NavPoint.cs
index 473d7fa..a70fd61 100755
--- a/Assets/Scripts/Misc/NavPoint.cs
+++ b/Assets/Scripts/Misc/NavPoint.cs
@@ -17,21 +17,12 @@ public class NavPoint : MonoBehaviour
     public NavPointType navType = NavPointType.Direction;
 
     [HideInInspector]
-<<<<<<< HEAD
     public int PointId = 0;
-=======
-    public int PointId;
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
     public float DeathAttr = 0;
     public List<Vector3> EnemiesSeen = new List<Vector3>();
 
     private void Start()
     {
-<<<<<<< HEAD
         FlagDistance = (GameObject.FindGameObjectWithTag("Flag").transform.position - Position).magnitude;
-=======
-        PointId = GetInstanceID();
-        FlagDistance = (GameObject.FindGameObjectWithTag("Flag").transform.position - position).magnitude;               
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
     }
 }
-- 
2.49.0


From 2caf658d190d294de644e3042211c96566daa261 Mon Sep 17 00:00:00 2001
From: Krazerleo <krazerleo@gmail.com>
Date: Thu, 5 May 2022 18:11:57 +0700
Subject: [PATCH 10/17] "update unresolved conflicts"

---
 Assets/Scripts/Character/CharacterCondition.cs |  1 -
 Assets/Scripts/Character/MovementController.cs | 10 ----------
 Assets/Scripts/Managers/MapManager.cs          |  2 --
 3 files changed, 13 deletions(-)

diff --git a/Assets/Scripts/Character/CharacterCondition.cs b/Assets/Scripts/Character/CharacterCondition.cs
index 8d7d5c5..2124ddb 100755
--- a/Assets/Scripts/Character/CharacterCondition.cs
+++ b/Assets/Scripts/Character/CharacterCondition.cs
@@ -9,7 +9,6 @@ public class CharacterCondition
     public event Action<int> OnChangeAmmunitionEvent;
 
     private int health;
-
     public int HealthPoints 
     { 
         get 
diff --git a/Assets/Scripts/Character/MovementController.cs b/Assets/Scripts/Character/MovementController.cs
index b1849ce..70341bb 100644
--- a/Assets/Scripts/Character/MovementController.cs
+++ b/Assets/Scripts/Character/MovementController.cs
@@ -36,16 +36,6 @@ public class MovementController : MonoBehaviour
     {
         FlagDistance = (flag.transform.position - gameObject.transform.position).magnitude;
     }
-
-    public void MoveToPointById(int id)
-    {
-        if (!navPoints.ContainsKey(id))
-        {
-            Debug.LogWarning("PIDOR");
-            return;
-        }
-        goToNextNavPoint(navPoints[id]);
-    }
     
     public void MoveToRandomPoint()
     {
diff --git a/Assets/Scripts/Managers/MapManager.cs b/Assets/Scripts/Managers/MapManager.cs
index b28dcba..58e5c77 100755
--- a/Assets/Scripts/Managers/MapManager.cs
+++ b/Assets/Scripts/Managers/MapManager.cs
@@ -5,8 +5,6 @@ public class MapManager : MonoBehaviour
 {
     private static MapManager instance;
     public static MapManager Instance => instance;
-    private static List<NavPoint> navPoints = new List<NavPoint>();
-    private static Dictionary<int, NavPoint> iDToNavPoint = new Dictionary<int, NavPoint>();
     public static List<NavPoint> NavPoints { get => navPoints; private set => navPoints = value; }
     public static Dictionary<int, NavPoint> IDToNavPoint { get => iDToNavPoint; private set => iDToNavPoint = value; }
 
-- 
2.49.0


From b8e8e74ab448c9126b0c22b32ed7b1d25507d97d Mon Sep 17 00:00:00 2001
From: Krazerleo <krazerleo@gmail.com>
Date: Thu, 5 May 2022 20:45:26 +0700
Subject: [PATCH 11/17] Fixes in Map Manager

---
 Assets/Scripts/Managers/GameManager.cs        |  8 ++++----
 Assets/Scripts/Managers/GlobalEventManager.cs | 12 ++++++------
 Assets/Scripts/Managers/MapManager.cs         |  5 +++--
 Assets/Scripts/Statistics/Logger.cs           | 14 +++++++-------
 Assets/Scripts/Statistics/StatisticManager.cs |  4 ++--
 5 files changed, 22 insertions(+), 21 deletions(-)

diff --git a/Assets/Scripts/Managers/GameManager.cs b/Assets/Scripts/Managers/GameManager.cs
index b6467d8..9299c9b 100755
--- a/Assets/Scripts/Managers/GameManager.cs
+++ b/Assets/Scripts/Managers/GameManager.cs
@@ -25,8 +25,8 @@ public class GameManager : MonoBehaviour
     {
         Academy.Instance.OnEnvironmentReset += ResetScene;
 
-        GlobalEventManager.onCaptureFlag += FlagCaptured;
-        GlobalEventManager.onTimeLeft += TimeOut;
+        GlobalEventManager.OnCaptureFlag += FlagCaptured;
+        GlobalEventManager.OnTimeLeft += TimeOut;
 
         var agents = GameObject.FindObjectsOfType<Agent>();
         foreach (var item in agents)
@@ -138,8 +138,8 @@ public class GameManager : MonoBehaviour
 
     private void OnDestroy()
     {
-        GlobalEventManager.onCaptureFlag -= FlagCaptured;
-        GlobalEventManager.onTimeLeft -= TimeOut;
+        GlobalEventManager.OnCaptureFlag -= FlagCaptured;
+        GlobalEventManager.OnTimeLeft -= TimeOut;
     }
 
     public static event Action OnResetScene;
diff --git a/Assets/Scripts/Managers/GlobalEventManager.cs b/Assets/Scripts/Managers/GlobalEventManager.cs
index e0461af..753b7d2 100755
--- a/Assets/Scripts/Managers/GlobalEventManager.cs
+++ b/Assets/Scripts/Managers/GlobalEventManager.cs
@@ -2,18 +2,18 @@
 
 public class GlobalEventManager
 {
-    public static event Action<Team> onCaptureFlag;
+    public static event Action<Team> OnCaptureFlag;
 
     public static void SendCaptureFlag(Team team)
     {
-        onCaptureFlag?.Invoke(team);
-        onCaptureFlag = null;
+        OnCaptureFlag?.Invoke(team);
+        OnCaptureFlag = null;
     }
 
-    public static event Action onTimeLeft;
+    public static event Action OnTimeLeft;
     public static void SendTimeout()
     {
-        onTimeLeft?.Invoke();
-        onTimeLeft = null;
+        OnTimeLeft?.Invoke();
+        OnTimeLeft = null;
     }
 }
diff --git a/Assets/Scripts/Managers/MapManager.cs b/Assets/Scripts/Managers/MapManager.cs
index 58e5c77..14d70af 100755
--- a/Assets/Scripts/Managers/MapManager.cs
+++ b/Assets/Scripts/Managers/MapManager.cs
@@ -5,8 +5,9 @@ public class MapManager : MonoBehaviour
 {
     private static MapManager instance;
     public static MapManager Instance => instance;
-    public static List<NavPoint> NavPoints { get => navPoints; private set => navPoints = value; }
-    public static Dictionary<int, NavPoint> IDToNavPoint { get => iDToNavPoint; private set => iDToNavPoint = value; }
+    [SerializeField] static List<NavPoint> _navPoints;
+    public static List<NavPoint> NavPoints { get => _navPoints; private set => _navPoints = value; }
+    public static Dictionary<int, NavPoint> IDToNavPoint {get; private set; }
 
     private void Awake()
     {
diff --git a/Assets/Scripts/Statistics/Logger.cs b/Assets/Scripts/Statistics/Logger.cs
index e293d25..df7ce3d 100644
--- a/Assets/Scripts/Statistics/Logger.cs
+++ b/Assets/Scripts/Statistics/Logger.cs
@@ -3,17 +3,17 @@ using UnityEngine;
 
 public class Logger
 {
-    private const string directory = "/Logs/";
-    private const string baseName = "Log#";
+    private const string Directory = "/Logs/";
+    private const string BaseName = "Log#";
 
     public static void SaveLog<T>(T objToSerialize)
     {
-        string dir = Application.persistentDataPath + directory;
-        if (!Directory.Exists(dir))
-            Directory.CreateDirectory(dir);
+        var dir = Application.persistentDataPath + Directory;
+        if (!System.IO.Directory.Exists(dir))
+            System.IO.Directory.CreateDirectory(dir);
 
-        var logName = baseName + (Directory.GetFiles(dir).Length + 1).ToString();
-        string json = JsonUtility.ToJson(objToSerialize);
+        var logName = BaseName + (System.IO.Directory.GetFiles(dir).Length + 1).ToString();
+        var json = JsonUtility.ToJson(objToSerialize);
         File.WriteAllText(dir + logName, json);
     }
 }
\ No newline at end of file
diff --git a/Assets/Scripts/Statistics/StatisticManager.cs b/Assets/Scripts/Statistics/StatisticManager.cs
index 399e7fb..a9c4883 100644
--- a/Assets/Scripts/Statistics/StatisticManager.cs
+++ b/Assets/Scripts/Statistics/StatisticManager.cs
@@ -19,8 +19,8 @@ public class StatisticManager : MonoBehaviour
         foreach (var npc in GameObject.FindObjectsOfType<NPC>())
             npc.OnDamageRecieved += RegisterDamage;
 
-        GlobalEventManager.onCaptureFlag += RegisterWin;
-        GlobalEventManager.onTimeLeft += RegisterTimeOut;
+        GlobalEventManager.OnCaptureFlag += RegisterWin;
+        GlobalEventManager.OnTimeLeft += RegisterTimeOut;
     }
 
     private void RegisterDamage(int damage, Team team)
-- 
2.49.0


From 3908334f656b248dc1f2eb4fe5a7918685dd472e Mon Sep 17 00:00:00 2001
From: Krazerleo <krazerleo@gmail.com>
Date: Sat, 7 May 2022 21:45:28 +0700
Subject: [PATCH 12/17] Finally fixed

---
 Assets/ML-Agents/Timers/Greatest_map_ever_timers.json.meta | 7 -------
 ProjectSettings/ProjectVersion.txt                         | 2 --
 2 files changed, 9 deletions(-)
 delete mode 100755 Assets/ML-Agents/Timers/Greatest_map_ever_timers.json.meta
 delete mode 100755 ProjectSettings/ProjectVersion.txt

diff --git a/Assets/ML-Agents/Timers/Greatest_map_ever_timers.json.meta b/Assets/ML-Agents/Timers/Greatest_map_ever_timers.json.meta
deleted file mode 100755
index 1400775..0000000
--- a/Assets/ML-Agents/Timers/Greatest_map_ever_timers.json.meta
+++ /dev/null
@@ -1,7 +0,0 @@
-fileFormatVersion: 2
-guid: 80ef0d75029e25243857877facd14d75
-TextScriptImporter:
-  externalObjects: {}
-  userData: 
-  assetBundleName: 
-  assetBundleVariant: 
diff --git a/ProjectSettings/ProjectVersion.txt b/ProjectSettings/ProjectVersion.txt
deleted file mode 100755
index bf8f7eb..0000000
--- a/ProjectSettings/ProjectVersion.txt
+++ /dev/null
@@ -1,2 +0,0 @@
-m_EditorVersion: 2019.4.36f1
-m_EditorVersionWithRevision: 2019.4.36f1 (660c164b2fc5)
-- 
2.49.0


From 87568c4077fa09ca9106d5a544aa848f754a6d41 Mon Sep 17 00:00:00 2001
From: Krazerleo <krazerleo@gmail.com>
Date: Sat, 7 May 2022 22:00:19 +0700
Subject: [PATCH 13/17] added scene

---
 .../Greatest_map_ever/Greatest_map_ever.unity | 212 +-----------------
 1 file changed, 6 insertions(+), 206 deletions(-)
 mode change 100755 => 100644 Assets/Scenes/Greatest_map_ever/Greatest_map_ever.unity

diff --git a/Assets/Scenes/Greatest_map_ever/Greatest_map_ever.unity b/Assets/Scenes/Greatest_map_ever/Greatest_map_ever.unity
old mode 100755
new mode 100644
index 19be7a0..1b0de26
--- a/Assets/Scenes/Greatest_map_ever/Greatest_map_ever.unity
+++ b/Assets/Scenes/Greatest_map_ever/Greatest_map_ever.unity
@@ -38,7 +38,7 @@ RenderSettings:
   m_ReflectionIntensity: 1
   m_CustomReflection: {fileID: 0}
   m_Sun: {fileID: 705507994}
-  m_IndirectSpecularColor: {r: 0.44657815, g: 0.49641192, b: 0.57481617, a: 1}
+  m_IndirectSpecularColor: {r: 0.44657898, g: 0.4964133, b: 0.5748178, a: 1}
   m_UseRadianceAmbientProbe: 0
 --- !u!157 &3
 LightmapSettings:
@@ -259,10 +259,7 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-<<<<<<< HEAD
   navType: 1
-=======
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
   PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
@@ -358,10 +355,7 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-<<<<<<< HEAD
   navType: 1
-=======
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
   PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
@@ -1130,7 +1124,7 @@ Transform:
   m_PrefabAsset: {fileID: 0}
   m_GameObject: {fileID: 705507993}
   m_LocalRotation: {x: 0.40821788, y: -0.23456968, z: 0.10938163, w: 0.8754261}
-  m_LocalPosition: {x: 81.2, y: 64.1, z: -12.4}
+  m_LocalPosition: {x: 77.5, y: 55.2, z: -5.9}
   m_LocalScale: {x: 1, y: 1, z: 1}
   m_Children: []
   m_Father: {fileID: 0}
@@ -1261,12 +1255,6 @@ Transform:
   m_Father: {fileID: 671439045}
   m_RootOrder: 1
   m_LocalEulerAnglesHint: {x: 0, y: 90, z: 0}
---- !u!4 &868386701 stripped
-Transform:
-  m_CorrespondingSourceObject: {fileID: 4528203470625763689, guid: 99bb17a23a489624baeaf337f91a4f84,
-    type: 3}
-  m_PrefabInstance: {fileID: 1809549197}
-  m_PrefabAsset: {fileID: 0}
 --- !u!1 &884498019
 GameObject:
   m_ObjectHideFlags: 0
@@ -1300,80 +1288,6 @@ Transform:
   m_Father: {fileID: 782729761}
   m_RootOrder: 0
   m_LocalEulerAnglesHint: {x: 0, y: 0, z: 0}
---- !u!1001 &1061293905
-PrefabInstance:
-  m_ObjectHideFlags: 0
-  serializedVersion: 2
-  m_Modification:
-    m_TransformParent: {fileID: 868386701}
-    m_Modifications:
-    - target: {fileID: 2836004473841745626, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_Controller
-      value: 
-      objectReference: {fileID: 9100000, guid: 95a44e56d04c7d248ba723eda9611c51, type: 2}
-    - target: {fileID: 2836004475591197023, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_RootOrder
-      value: 2
-      objectReference: {fileID: 0}
-    - target: {fileID: 2836004475591197023, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_LocalPosition.x
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 2836004475591197023, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_LocalPosition.y
-      value: -1
-      objectReference: {fileID: 0}
-    - target: {fileID: 2836004475591197023, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_LocalPosition.z
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 2836004475591197023, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_LocalRotation.w
-      value: 1
-      objectReference: {fileID: 0}
-    - target: {fileID: 2836004475591197023, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_LocalRotation.x
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 2836004475591197023, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_LocalRotation.y
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 2836004475591197023, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_LocalRotation.z
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 2836004475591197023, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_LocalEulerAnglesHint.x
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 2836004475591197023, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_LocalEulerAnglesHint.y
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 2836004475591197023, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_LocalEulerAnglesHint.z
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 2836004475591695743, guid: f432554f564e69242897607d34218939,
-        type: 3}
-      propertyPath: m_Name
-      value: DragonFucker
-      objectReference: {fileID: 0}
-    m_RemovedComponents: []
-  m_SourcePrefab: {fileID: 100100000, guid: f432554f564e69242897607d34218939, type: 3}
 --- !u!1 &1116745543
 GameObject:
   m_ObjectHideFlags: 0
@@ -1405,10 +1319,7 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-<<<<<<< HEAD
   navType: 1
-=======
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
   PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
@@ -1840,7 +1751,7 @@ Transform:
   m_PrefabAsset: {fileID: 0}
   m_GameObject: {fileID: 1345085340}
   m_LocalRotation: {x: -0, y: -0, z: -0, w: 1}
-  m_LocalPosition: {x: 21.51, y: -5.782543, z: -10.46}
+  m_LocalPosition: {x: 47.446796, y: -5.782543, z: -25.400002}
   m_LocalScale: {x: 1.4367, y: 1.4367, z: 1.4367}
   m_Children: []
   m_Father: {fileID: 2060099472}
@@ -1858,10 +1769,7 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-<<<<<<< HEAD
   navType: 1
-=======
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
   PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
@@ -1957,10 +1865,7 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-<<<<<<< HEAD
   navType: 1
-=======
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
   PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
@@ -2182,10 +2087,7 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-<<<<<<< HEAD
   navType: 1
-=======
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
   PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
@@ -2236,105 +2138,6 @@ MeshFilter:
   m_PrefabAsset: {fileID: 0}
   m_GameObject: {fileID: 1663305221}
   m_Mesh: {fileID: 10206, guid: 0000000000000000e000000000000000, type: 0}
---- !u!1001 &1809549197
-PrefabInstance:
-  m_ObjectHideFlags: 0
-  serializedVersion: 2
-  m_Modification:
-    m_TransformParent: {fileID: 0}
-    m_Modifications:
-    - target: {fileID: -4942972567661207728, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_Enabled
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763689, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_RootOrder
-      value: 6
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763689, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_LocalPosition.x
-      value: 42.23
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763689, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_LocalPosition.y
-      value: 1
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763689, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_LocalPosition.z
-      value: -15.91
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763689, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_LocalRotation.w
-      value: 1
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763689, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_LocalRotation.x
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763689, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_LocalRotation.y
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763689, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_LocalRotation.z
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763689, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_LocalEulerAnglesHint.x
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763689, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_LocalEulerAnglesHint.y
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763689, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_LocalEulerAnglesHint.z
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763699, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_Name
-      value: Player
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763700, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: playerStance
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763700, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: playerCrouchStance.CameraHeight
-      value: 0.3
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203470625763703, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_Enabled
-      value: 1
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203471293941515, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_LocalPosition.x
-      value: 0
-      objectReference: {fileID: 0}
-    - target: {fileID: 4528203471293941515, guid: 99bb17a23a489624baeaf337f91a4f84,
-        type: 3}
-      propertyPath: m_LocalPosition.z
-      value: -2.417
-      objectReference: {fileID: 0}
-    m_RemovedComponents: []
-  m_SourcePrefab: {fileID: 100100000, guid: 99bb17a23a489624baeaf337f91a4f84, type: 3}
 --- !u!1 &1858987083
 GameObject:
   m_ObjectHideFlags: 0
@@ -2459,9 +2262,9 @@ MonoBehaviour:
   - {fileID: 140697606}
   spawnPointsForAttackersTeam:
   - {fileID: 2004854094}
-  AIPrefab: {fileID: 2988578997639256874, guid: b016874eb34cc084aa4359f0bbec50e1,
+  AIPrefab: {fileID: 6171680323948707524, guid: b016874eb34cc084aa4359f0bbec50e1,
     type: 3}
-  PlayerPrefab: {fileID: 5245491127989480125, guid: 99bb17a23a489624baeaf337f91a4f84,
+  PlayerPrefab: {fileID: 5245491127989480125, guid: 80f6c1c85e5daed4c96c70205ed5503d,
     type: 3}
 --- !u!114 &1858987091
 MonoBehaviour:
@@ -2728,10 +2531,7 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: 6a2d29bfc31a8cf4e831e3bb80720414, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-<<<<<<< HEAD
   navType: 1
-=======
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
   PointId: 0
   DeathAttr: 0
   EnemiesSeen: []
@@ -2820,7 +2620,7 @@ Transform:
   m_PrefabAsset: {fileID: 0}
   m_GameObject: {fileID: 2060099471}
   m_LocalRotation: {x: -0, y: -0, z: -0, w: 1}
-  m_LocalPosition: {x: 27.79, y: 7.782543, z: -15}
+  m_LocalPosition: {x: 19.553204, y: 7.782543, z: -15}
   m_LocalScale: {x: 1, y: 1, z: 1}
   m_Children:
   - {fileID: 1345085341}
-- 
2.49.0


From f9a3b06ec37f5b9e18b1c6a89fa972fbb20116a5 Mon Sep 17 00:00:00 2001
From: Krazerleo <krazerleo@gmail.com>
Date: Sat, 7 May 2022 22:02:15 +0700
Subject: [PATCH 14/17] files add

---
 Assets/Prefabs/Bot.prefab                     | 237 +++++++++++-------
 Assets/Prefabs/Player.prefab                  | 150 ++++++-----
 .../Scripts/Character/MovementController.cs   |  21 +-
 Assets/Scripts/Character/NPC.cs               |  26 +-
 Assets/Scripts/Managers/GameManager.cs        |   1 -
 Assets/Scripts/Managers/MapManager.cs         |  30 +--
 .../Statistics/StatisticManager.cs.meta       |   2 +-
 Assets/Settings/Game Settings.asset           |  27 --
 8 files changed, 281 insertions(+), 213 deletions(-)

diff --git a/Assets/Prefabs/Bot.prefab b/Assets/Prefabs/Bot.prefab
index 0af3f18..5b177e0 100755
--- a/Assets/Prefabs/Bot.prefab
+++ b/Assets/Prefabs/Bot.prefab
@@ -1,6 +1,6 @@
 %YAML 1.1
 %TAG !u! tag:unity3d.com,2011:
---- !u!1 &2988578997639256874
+--- !u!1 &6171680323407988095
 GameObject:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
@@ -8,52 +8,85 @@ GameObject:
   m_PrefabAsset: {fileID: 0}
   serializedVersion: 6
   m_Component:
-  - component: {fileID: 2988578997639256870}
-  - component: {fileID: 2988578997639256869}
-  - component: {fileID: 2988578997639256868}
-  - component: {fileID: 2988578997639256875}
-  - component: {fileID: 5447337162552783061}
-  - component: {fileID: 7805954453358028498}
-  - component: {fileID: 2676446634235362783}
-  - component: {fileID: 8656710265340117963}
-  - component: {fileID: 778652956973742106}
-  - component: {fileID: 1208561866453126566}
-  m_Layer: 0
-  m_Name: Bot
+  - component: {fileID: 6171680323407988094}
+  m_Layer: 8
+  m_Name: FeetTransform
   m_TagString: Untagged
   m_Icon: {fileID: 0}
   m_NavMeshLayer: 0
   m_StaticEditorFlags: 0
   m_IsActive: 1
---- !u!4 &2988578997639256870
+--- !u!4 &6171680323407988094
 Transform:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
-  m_GameObject: {fileID: 2988578997639256874}
+  m_GameObject: {fileID: 6171680323407988095}
   m_LocalRotation: {x: 0, y: 0, z: 0, w: 1}
-  m_LocalPosition: {x: 28.310326, y: 13.98, z: 46.45}
+  m_LocalPosition: {x: 0, y: -1, z: 0}
   m_LocalScale: {x: 1, y: 1, z: 1}
   m_Children: []
-  m_Father: {fileID: 0}
+  m_Father: {fileID: 6171680323948707550}
   m_RootOrder: 0
   m_LocalEulerAnglesHint: {x: 0, y: 0, z: 0}
---- !u!33 &2988578997639256869
+--- !u!1 &6171680323948707524
+GameObject:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  serializedVersion: 6
+  m_Component:
+  - component: {fileID: 6171680323948707550}
+  - component: {fileID: 6171680323948707551}
+  - component: {fileID: 5770895893828047079}
+  - component: {fileID: 6171680323948707521}
+  - component: {fileID: 6171680323948707549}
+  - component: {fileID: 6171680323948707520}
+  - component: {fileID: 8774702625908438859}
+  - component: {fileID: 6521536090983603910}
+  - component: {fileID: 6133354754598649724}
+  - component: {fileID: 2756943273076691504}
+  - component: {fileID: 6638271233700792696}
+  - component: {fileID: 5988018701276467001}
+  m_Layer: 8
+  m_Name: Bot
+  m_TagString: Defender
+  m_Icon: {fileID: 0}
+  m_NavMeshLayer: 0
+  m_StaticEditorFlags: 0
+  m_IsActive: 1
+--- !u!4 &6171680323948707550
+Transform:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  m_GameObject: {fileID: 6171680323948707524}
+  m_LocalRotation: {x: 0, y: 0.7071068, z: 0, w: 0.7071068}
+  m_LocalPosition: {x: 50.67923, y: 1.16, z: -22.57}
+  m_LocalScale: {x: 1, y: 1, z: 1}
+  m_Children:
+  - {fileID: 6171680323407988094}
+  m_Father: {fileID: 0}
+  m_RootOrder: 0
+  m_LocalEulerAnglesHint: {x: 0, y: 90, z: 0}
+--- !u!33 &6171680323948707551
 MeshFilter:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
-  m_GameObject: {fileID: 2988578997639256874}
+  m_GameObject: {fileID: 6171680323948707524}
   m_Mesh: {fileID: 10208, guid: 0000000000000000e000000000000000, type: 0}
---- !u!23 &2988578997639256868
+--- !u!23 &5770895893828047079
 MeshRenderer:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
-  m_GameObject: {fileID: 2988578997639256874}
+  m_GameObject: {fileID: 6171680323948707524}
   m_Enabled: 1
   m_CastShadows: 1
   m_ReceiveShadows: 1
@@ -86,94 +119,58 @@ MeshRenderer:
   m_SortingLayerID: 0
   m_SortingLayer: 0
   m_SortingOrder: 0
---- !u!136 &2988578997639256875
+--- !u!136 &6171680323948707521
 CapsuleCollider:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
-  m_GameObject: {fileID: 2988578997639256874}
+  m_GameObject: {fileID: 6171680323948707524}
   m_Material: {fileID: 0}
   m_IsTrigger: 0
-  m_Enabled: 1
-  m_Radius: 0.5
+  m_Enabled: 0
+  m_Radius: 0.3
   m_Height: 2
   m_Direction: 1
   m_Center: {x: 0, y: 0, z: 0}
---- !u!114 &5447337162552783061
-MonoBehaviour:
+--- !u!136 &6171680323948707549
+CapsuleCollider:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
-  m_GameObject: {fileID: 2988578997639256874}
-  m_Enabled: 1
-  m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: a6f2a081cfc8c4b4bb6864331109d147, type: 3}
-  m_Name: 
-  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 0
-  hasUpgradedFromAgentParameters: 1
-  MaxStep: 100
---- !u!114 &7805954453358028498
-MonoBehaviour:
+  m_GameObject: {fileID: 6171680323948707524}
+  m_Material: {fileID: 0}
+  m_IsTrigger: 0
+  m_Enabled: 0
+  m_Radius: 0.3
+  m_Height: 1.3
+  m_Direction: 1
+  m_Center: {x: 0, y: -0.35, z: 0}
+--- !u!136 &6171680323948707520
+CapsuleCollider:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
-  m_GameObject: {fileID: 2988578997639256874}
+  m_GameObject: {fileID: 6171680323948707524}
+  m_Material: {fileID: 0}
+  m_IsTrigger: 0
   m_Enabled: 1
-  m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 3a5c9d521e5ef4759a8246a07d52221e, type: 3}
-  m_Name: 
-  m_EditorClassIdentifier: 
-  DecisionPeriod: 1
-  TakeActionsBetweenDecisions: 1
---- !u!114 &2676446634235362783
-MonoBehaviour:
-  m_ObjectHideFlags: 0
-  m_CorrespondingSourceObject: {fileID: 0}
-  m_PrefabInstance: {fileID: 0}
-  m_PrefabAsset: {fileID: 0}
-  m_GameObject: {fileID: 2988578997639256874}
-  m_Enabled: 1
-  m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 5d1c4e0b1822b495aa52bc52839ecb30, type: 3}
-  m_Name: 
-  m_EditorClassIdentifier: 
-  m_BrainParameters:
-<<<<<<< HEAD
-    VectorObservationSize: 9
-=======
-    VectorObservationSize: 4
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
-    NumStackedVectorObservations: 1
-    m_ActionSpec:
-      m_NumContinuousActions: 0
-      BranchSizes: 0200000002000000
-    VectorActionSize: 0200000002000000
-    VectorActionDescriptions: []
-    VectorActionSpaceType: 0
-    hasUpgradedBrainParametersWithActionSpec: 1
-  m_Model: {fileID: 0}
-  m_InferenceDevice: 0
-  m_BehaviorType: 0
-  m_BehaviorName: npc
-  TeamId: 0
-  m_UseChildSensors: 1
-  m_UseChildActuators: 1
-  m_ObservableAttributeHandling: 0
---- !u!195 &8656710265340117963
+  m_Radius: 0.3
+  m_Height: 0.8
+  m_Direction: 1
+  m_Center: {x: 0, y: -0.6, z: 0}
+--- !u!195 &8774702625908438859
 NavMeshAgent:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
-  m_GameObject: {fileID: 2988578997639256874}
+  m_GameObject: {fileID: 6171680323948707524}
   m_Enabled: 1
   m_AgentTypeID: 0
-  m_Radius: 0.5
+  m_Radius: 0.5000001
   m_Speed: 3.5
   m_Acceleration: 8
   avoidancePriority: 50
@@ -186,32 +183,92 @@ NavMeshAgent:
   m_BaseOffset: 1
   m_WalkableMask: 4294967295
   m_ObstacleAvoidanceType: 4
---- !u!114 &778652956973742106
+--- !u!114 &6521536090983603910
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
-  m_GameObject: {fileID: 2988578997639256874}
+  m_GameObject: {fileID: 6171680323948707524}
   m_Enabled: 1
   m_EditorHideFlags: 0
   m_Script: {fileID: 11500000, guid: d3ebcf807a37f344998fd648dfc9376d, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-  navMeshAgent: {fileID: 8656710265340117963}
+  navMeshAgent: {fileID: 8774702625908438859}
   flag: {fileID: 6818223691859422291, guid: 1685c1d9ce4ab174f95c646b1826010b, type: 3}
---- !u!114 &1208561866453126566
+--- !u!114 &6133354754598649724
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
-  m_GameObject: {fileID: 2988578997639256874}
+  m_GameObject: {fileID: 6171680323948707524}
   m_Enabled: 1
   m_EditorHideFlags: 0
   m_Script: {fileID: 11500000, guid: dd8012d5925524537b27131fef517017, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
   m_SensorName: BufferSensor
-  m_ObservableSize: 5
-  m_MaxNumObservables: 5
+  m_ObservableSize: 0
+  m_MaxNumObservables: 0
+--- !u!114 &2756943273076691504
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  m_GameObject: {fileID: 6171680323948707524}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 5d1c4e0b1822b495aa52bc52839ecb30, type: 3}
+  m_Name: 
+  m_EditorClassIdentifier: 
+  m_BrainParameters:
+    VectorObservationSize: 1
+    NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 0
+      BranchSizes: 01000000
+    VectorActionSize: 01000000
+    VectorActionDescriptions: []
+    VectorActionSpaceType: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 0}
+  m_InferenceDevice: 0
+  m_BehaviorType: 0
+  m_BehaviorName: My Behavior
+  TeamId: 0
+  m_UseChildSensors: 1
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 0
+--- !u!114 &6638271233700792696
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  m_GameObject: {fileID: 6171680323948707524}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: a6f2a081cfc8c4b4bb6864331109d147, type: 3}
+  m_Name: 
+  m_EditorClassIdentifier: 
+  agentParameters:
+    maxStep: 0
+  hasUpgradedFromAgentParameters: 1
+  MaxStep: 100
+--- !u!114 &5988018701276467001
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  m_GameObject: {fileID: 6171680323948707524}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: d3ebcf807a37f344998fd648dfc9376d, type: 3}
+  m_Name: 
+  m_EditorClassIdentifier: 
+  navMeshAgent: {fileID: 8774702625908438859}
+  flag: {fileID: 6818223691859422291, guid: 1685c1d9ce4ab174f95c646b1826010b, type: 3}
diff --git a/Assets/Prefabs/Player.prefab b/Assets/Prefabs/Player.prefab
index c798d61..31319fc 100755
--- a/Assets/Prefabs/Player.prefab
+++ b/Assets/Prefabs/Player.prefab
@@ -14,9 +14,12 @@ GameObject:
   - component: {fileID: 5245491127989480120}
   - component: {fileID: 5245491127989480100}
   - component: {fileID: 5245491127989480121}
-  - component: {fileID: 5245491127989480123}
-  - component: {fileID: 5245491127989480122}
-  - component: {fileID: 4890899368932544690}
+  - component: {fileID: 7254047075221496626}
+  - component: {fileID: 5182704636738128575}
+  - component: {fileID: 5220658550450318085}
+  - component: {fileID: 4264677542023120457}
+  - component: {fileID: 4715950961444674817}
+  - component: {fileID: 5653209015090846528}
   m_Layer: 8
   m_Name: Player
   m_TagString: Defender
@@ -129,25 +132,29 @@ CapsuleCollider:
   m_Height: 0.8
   m_Direction: 1
   m_Center: {x: 0, y: -0.6, z: 0}
---- !u!143 &5245491127989480123
-CharacterController:
+--- !u!195 &7254047075221496626
+NavMeshAgent:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
   m_PrefabInstance: {fileID: 0}
   m_PrefabAsset: {fileID: 0}
   m_GameObject: {fileID: 5245491127989480125}
-  m_Material: {fileID: 0}
-  m_IsTrigger: 0
   m_Enabled: 1
-  serializedVersion: 2
+  m_AgentTypeID: 0
+  m_Radius: 0.5000001
+  m_Speed: 3.5
+  m_Acceleration: 8
+  avoidancePriority: 50
+  m_AngularSpeed: 120
+  m_StoppingDistance: 0
+  m_AutoTraverseOffMeshLink: 1
+  m_AutoBraking: 1
+  m_AutoRepath: 1
   m_Height: 2
-  m_Radius: 0.5
-  m_SlopeLimit: 45
-  m_StepOffset: 0.3
-  m_SkinWidth: 0.08
-  m_MinMoveDistance: 0.001
-  m_Center: {x: 0, y: 0, z: 0}
---- !u!114 &5245491127989480122
+  m_BaseOffset: 1
+  m_WalkableMask: 4294967295
+  m_ObstacleAvoidanceType: 4
+--- !u!114 &5182704636738128575
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
@@ -156,52 +163,12 @@ MonoBehaviour:
   m_GameObject: {fileID: 5245491127989480125}
   m_Enabled: 1
   m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 9826297ef4d853741b2af768441ec7f7, type: 3}
+  m_Script: {fileID: 11500000, guid: d3ebcf807a37f344998fd648dfc9376d, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
-  input_View: {x: 0, y: 0}
-  cameraHolder: {fileID: 5245491129196666053}
-  feetTransform: {fileID: 5245491129603592455}
-  playerSettings:
-    ViewXSensetivity: 20
-    ViewYSensetivity: 20
-    ViewXInverted: 0
-    ViewYInverted: 0
-    SprintingHold: 0
-    MovementSmoothing: 0
-    RunningForwardSpeed: 10
-    RunningStrafeSpeed: 6
-    WalkingForwardSpeed: 4
-    WalkingBackwardSpeed: 2
-    WalkingStrafeSpeed: 3
-    JumpingHeight: 6
-    JumpingFalloff: 1
-    FallingSmoothing: 0
-    SpeedEffector: 1
-    CrouchSpeedEffector: 0
-    ProneSpeedEffector: 0
-    FallingSpeedEffector: 0
-  ViewClampYMin: -70
-  ViewClampYMax: 80
-  playerMask:
-    serializedVersion: 2
-    m_Bits: 55
-  gravityAmount: 0.05
-  gravityMin: -3
-  jumpingForce: {x: 0, y: 0, z: 0}
-  playerStance: 0
-  playerStanceSmoothing: 0.2
-  playerStandStance:
-    CameraHeight: 0.7
-    StanceCollider: {fileID: 5245491127989480120}
-  playerCrouchStance:
-    CameraHeight: 0.3
-    StanceCollider: {fileID: 5245491127989480100}
-  playerProneStance:
-    CameraHeight: -0.58
-    StanceCollider: {fileID: 5245491127989480121}
-  currentWeapon: {fileID: 8510909888689775086}
---- !u!114 &4890899368932544690
+  navMeshAgent: {fileID: 0}
+  flag: {fileID: 0}
+--- !u!114 &5220658550450318085
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_CorrespondingSourceObject: {fileID: 0}
@@ -210,9 +177,72 @@ MonoBehaviour:
   m_GameObject: {fileID: 5245491127989480125}
   m_Enabled: 1
   m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: a8c9a8e604d395c4ab9d03d28adc4982, type: 3}
+  m_Script: {fileID: 11500000, guid: dd8012d5925524537b27131fef517017, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+  m_SensorName: BufferSensor
+  m_ObservableSize: 0
+  m_MaxNumObservables: 0
+--- !u!114 &4264677542023120457
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  m_GameObject: {fileID: 5245491127989480125}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 5d1c4e0b1822b495aa52bc52839ecb30, type: 3}
+  m_Name: 
+  m_EditorClassIdentifier: 
+  m_BrainParameters:
+    VectorObservationSize: 1
+    NumStackedVectorObservations: 1
+    m_ActionSpec:
+      m_NumContinuousActions: 0
+      BranchSizes: 01000000
+    VectorActionSize: 01000000
+    VectorActionDescriptions: []
+    VectorActionSpaceType: 0
+    hasUpgradedBrainParametersWithActionSpec: 1
+  m_Model: {fileID: 0}
+  m_InferenceDevice: 0
+  m_BehaviorType: 0
+  m_BehaviorName: My Behavior
+  TeamId: 0
+  m_UseChildSensors: 1
+  m_UseChildActuators: 1
+  m_ObservableAttributeHandling: 0
+--- !u!114 &4715950961444674817
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  m_GameObject: {fileID: 5245491127989480125}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: a6f2a081cfc8c4b4bb6864331109d147, type: 3}
+  m_Name: 
+  m_EditorClassIdentifier: 
+  agentParameters:
+    maxStep: 0
+  hasUpgradedFromAgentParameters: 1
+  MaxStep: 0
+--- !u!114 &5653209015090846528
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  m_GameObject: {fileID: 5245491127989480125}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: d3ebcf807a37f344998fd648dfc9376d, type: 3}
+  m_Name: 
+  m_EditorClassIdentifier: 
+  navMeshAgent: {fileID: 7254047075221496626}
+  flag: {fileID: 0}
 --- !u!1 &5245491128202443531
 GameObject:
   m_ObjectHideFlags: 0
diff --git a/Assets/Scripts/Character/MovementController.cs b/Assets/Scripts/Character/MovementController.cs
index 70341bb..98df2aa 100644
--- a/Assets/Scripts/Character/MovementController.cs
+++ b/Assets/Scripts/Character/MovementController.cs
@@ -9,27 +9,28 @@ public class MovementController : MonoBehaviour
     public int PointStartID { get; set; }
     public int PointEndID { get; private set; }
     public float FlagDistance { get; private set; }
-    private const float updateFlagPositionDelay = 5;
-    private const float updateReachedDestinationDelay = 5;
+    private const float UpdateFlagPositionDelay = 5;
+    private const float UpdateReachedDestinationDelay = 5;
 
     [SerializeField] private NavMeshAgent navMeshAgent;
     [SerializeField] private GameObject flag;
     public float DistanceToGo { get; private set; }
     public float RemainingDistance => navMeshAgent.remainingDistance;
-    private Dictionary<int, NavPoint> idNavPointDict;
+    private Dictionary<int, NavPoint> _idNavPointDict;
 
 
     private void Awake()
     {
         navMeshAgent.speed = SettingsReader.Instance.GetSettings.MovementSpeed;
-        idNavPointDict = MapManager.IDToNavPoint;
-        InvokeRepeating(nameof(UpdateFlagPosition), 0, updateFlagPositionDelay);
-        InvokeRepeating(nameof(ReachedDestination), 0, updateReachedDestinationDelay);
+        _idNavPointDict = MapManager.Instance.IDToNavPoint;
+        InvokeRepeating(nameof(UpdateFlagPosition), 0, UpdateFlagPositionDelay);
+        InvokeRepeating(nameof(ReachedDestination), 0, UpdateReachedDestinationDelay);
     }
    
     private void OnDestroy()
     {
         CancelInvoke(nameof(UpdateFlagPosition));
+        CancelInvoke(nameof(ReachedDestination));
     }
 
     private void UpdateFlagPosition()
@@ -39,14 +40,14 @@ public class MovementController : MonoBehaviour
     
     public void MoveToRandomPoint()
     {
-        GoToNextNavPoint(MapManager.NavPoints[Random.Range(0, MapManager.NavPoints.Count)]);
+        GoToNextNavPoint(MapManager.Instance.NavPoints[Random.Range(0, MapManager.Instance.NavPoints.Count)]);
     }
 
     public List<NavPoint> GetPointsCandidate()
     {
-        return MapManager.NavPoints
+        return MapManager.Instance.NavPoints
             .Where(point => 
-            (idNavPointDict[PointStartID].Position - point.Position).magnitude < SettingsReader.Instance.GetSettings.MovementDistance)
+            (_idNavPointDict[PointStartID].Position - point.Position).magnitude < SettingsReader.Instance.GetSettings.MovementDistance)
             .ToList();
     }
 
@@ -62,7 +63,7 @@ public class MovementController : MonoBehaviour
     public void ReturnToStartPoint()
     {
         if (navMeshAgent.isStopped == true) navMeshAgent.isStopped = false;
-        navMeshAgent.SetDestination(idNavPointDict[PointStartID].Position);
+        navMeshAgent.SetDestination(_idNavPointDict[PointStartID].Position);
         PointEndID = PointStartID;
         PointStartID = -1;
     }
diff --git a/Assets/Scripts/Character/NPC.cs b/Assets/Scripts/Character/NPC.cs
index d47e377..36b76ec 100644
--- a/Assets/Scripts/Character/NPC.cs
+++ b/Assets/Scripts/Character/NPC.cs
@@ -5,7 +5,7 @@ using Unity.MLAgents.Actuators;
 using Unity.MLAgents.Sensors;
 using UnityEngine;
 
-[RequireComponent(typeof(MovementController),typeof(BufferSensor))]
+[RequireComponent(typeof(MovementController),typeof(BufferSensorComponent))]
 public class NPC : Agent, ICharacter
 {
     [HideInInspector]
@@ -47,12 +47,12 @@ public class NPC : Agent, ICharacter
 
         moveController = gameObject.GetComponent<MovementController>();
         bufferSensor = gameObject.GetComponent<BufferSensorComponent>();
-
+        
         flagZone = GameObject.FindObjectOfType<FlagZone>();
-        if (flagZone == null)
-            Debug.LogError("Flag Is Not Setted");
-
-        navPointIdDict = MapManager.IDToNavPoint;
+        if (flagZone is null)
+            Debug.LogError("Flag Is Not Set");
+        
+        navPointIdDict = MapManager.Instance.IDToNavPoint;
         if (navPointIdDict is null)
             Debug.LogError("Cant Find Nav Point Dictionary");
     }
@@ -64,12 +64,18 @@ public class NPC : Agent, ICharacter
 
     public override void OnEpisodeBegin()
     {
+        if (navPointIdDict is null)
+            Debug.LogError("Cant Find Nav Point Dictionary");
+        
         NpcState = DirectState;
         flagZone = GameObject.FindObjectOfType<FlagZone>();
     }
 
     public override void CollectObservations(VectorSensor sensor)
     {
+        navPointIdDict = MapManager.Instance.IDToNavPoint;
+        if (navPointIdDict is null)
+            Debug.LogError("Cant Find Nav Point Dictionary");
         var candidates = moveController.GetPointsCandidate();
 
         //common sensors
@@ -93,13 +99,14 @@ public class NPC : Agent, ICharacter
         //point sensors
         foreach (var point in candidates)
         {
+            var position = transform.position;
             bufferSensor.AppendObservation(new float[] {
                 point.DeathAttr,
                 (int)point.navType,
                 //4 flagEnemyDistance
-                GameManager.IsCloserToFlagFromNextNavPoint(point, transform.position).ToInt(),
+                GameManager.IsCloserToFlagFromNextNavPoint(point, position).ToInt(),
                 //5 EnemyVsNavPointDistance
-                GameManager.IsCloserToEnemyThanToNextNavPoint(point,transform.position, AgentCharacter.Team.GetOppositeTeam()).ToInt(),
+                GameManager.IsCloserToEnemyThanToNextNavPoint(point, position, AgentCharacter.Team.GetOppositeTeam()).ToInt(),
                 //6 Have been seen by enemy in this point               
                 GameManager.IsHaveSeenByEnemy(AgentCharacter.Team.GetOppositeTeam(),
                               point.Position).ToInt()
@@ -165,11 +172,12 @@ public class NPC : Agent, ICharacter
     }
 
     public event Action<int, Team> OnDamageRecieved;
-    public void GetDamage(float damage)
+    public void GetDamage(int damage)
     {
         AgentCharacter.LastTimeHit = TimeManager.Instance.CurrentTime;
         Condition.GiveHealth(-Mathf.RoundToInt(damage * (1 - Condition.ArmourPoints * 0.5f)));
         Condition.GiveArmour(-Mathf.RoundToInt(Mathf.Sqrt(damage) * 5));
+        OnDamageRecieved?.Invoke(damage, AgentCharacter.Team);
 
         if (Condition.HealthPoints < 0)
         {
diff --git a/Assets/Scripts/Managers/GameManager.cs b/Assets/Scripts/Managers/GameManager.cs
index 9299c9b..431fd81 100755
--- a/Assets/Scripts/Managers/GameManager.cs
+++ b/Assets/Scripts/Managers/GameManager.cs
@@ -24,7 +24,6 @@ public class GameManager : MonoBehaviour
     private void Start()
     {
         Academy.Instance.OnEnvironmentReset += ResetScene;
-
         GlobalEventManager.OnCaptureFlag += FlagCaptured;
         GlobalEventManager.OnTimeLeft += TimeOut;
 
diff --git a/Assets/Scripts/Managers/MapManager.cs b/Assets/Scripts/Managers/MapManager.cs
index 14d70af..97a0c75 100755
--- a/Assets/Scripts/Managers/MapManager.cs
+++ b/Assets/Scripts/Managers/MapManager.cs
@@ -3,36 +3,36 @@ using UnityEngine;
 
 public class MapManager : MonoBehaviour
 {
-    private static MapManager instance;
-    public static MapManager Instance => instance;
-    [SerializeField] static List<NavPoint> _navPoints;
-    public static List<NavPoint> NavPoints { get => _navPoints; private set => _navPoints = value; }
-    public static Dictionary<int, NavPoint> IDToNavPoint {get; private set; }
+    private static MapManager _instance;
+    public static MapManager Instance => _instance;
+    [SerializeField] private List<NavPoint> _navPoints;
+    public List<NavPoint> NavPoints { get => _navPoints; private set => _navPoints = value; }
+    public Dictionary<int, NavPoint> IDToNavPoint {get; private set; }
 
     private void Awake()
     {
-        if (instance is null)
-            instance = this;
+        if (_instance is null)
+            _instance = this;
         else
         {
             Destroy(gameObject);
             Debug.LogError("Only 1 Instance");
         }
-    }
-
-    private void Start()
-    {
+        
+        NavPoints = new List<NavPoint>();
         var navPointSet = GameObject.Find("NavPoint Set");
         var count = navPointSet.transform.childCount;
-        for (int i=0; i < count; i++)        
+        for (var i=0; i < count; i++)        
             NavPoints.Add(navPointSet.transform.GetChild(i)
                 .gameObject.GetComponent<NavPoint>());
-
+        print(NavPoints.Count);
         NavPointSetToID();
     }
+    
 
     private void NavPointSetToID()
     {
+        IDToNavPoint = new Dictionary<int, NavPoint>();
         int i = 0;
         foreach (var navPoint in NavPoints)
         {
@@ -45,8 +45,8 @@ public class MapManager : MonoBehaviour
     public static void AddDeathAttributeToPoints(int startPoint, int endPoint, 
         float allDistance, float remainingDistance)
     {
-        var startNavPoint = IDToNavPoint[startPoint];
-        var endNavPoint = IDToNavPoint[endPoint];
+        var startNavPoint = _instance.IDToNavPoint[startPoint];
+        var endNavPoint = _instance.IDToNavPoint[endPoint];
         float coef; 
         try
         {
diff --git a/Assets/Scripts/Statistics/StatisticManager.cs.meta b/Assets/Scripts/Statistics/StatisticManager.cs.meta
index 3a27c34..b8be93a 100644
--- a/Assets/Scripts/Statistics/StatisticManager.cs.meta
+++ b/Assets/Scripts/Statistics/StatisticManager.cs.meta
@@ -4,7 +4,7 @@ MonoImporter:
   externalObjects: {}
   serializedVersion: 2
   defaultReferences: []
-  executionOrder: 300
+  executionOrder: 0
   icon: {instanceID: 0}
   userData: 
   assetBundleName: 
diff --git a/Assets/Settings/Game Settings.asset b/Assets/Settings/Game Settings.asset
index ea0fad7..fc07488 100755
--- a/Assets/Settings/Game Settings.asset	
+++ b/Assets/Settings/Game Settings.asset	
@@ -12,7 +12,6 @@ MonoBehaviour:
   m_Script: {fileID: 11500000, guid: e2c47233b9062c84482336b145c6891b, type: 3}
   m_Name: Game Settings
   m_EditorClassIdentifier: 
-<<<<<<< HEAD
   IsTesting: 0
   TimeToWin: 15
   TimeOut: 3600
@@ -39,29 +38,3 @@ MonoBehaviour:
   DoDamageChanceInRunning: 30
   DoDamageChanceInCover: 25
   CrouchingCoefficient: 1.4
-=======
-  isTesting: 1
-  timeToWin: 5
-  timeOut: 1600
-  movementDistance: 50
-  movementSpeed: 3
-  defTeamAI: 0
-  atcTeamAI: 0
-  numOfDefenders: 1
-  numOfAttackers: 1
-  hasHumanDefender: 1
-  hasHumanAttacker: 0
-  healthPickupAmount: 50
-  armourPickupAmount: 50
-  ammunitionPickupAmount: 120
-  pickupsAmount: 2
-  maxHealth: 100
-  maxArmour: 100
-  maxAmmo: 31337
-  GetHitChanceInDirectPoint: 0
-  GetHitChanceInRunning: 0
-  GetHitChanceInCover: 0
-  DoDamageChanceInDirectPoint: 0
-  DoDamageChanceInRunning: 0
-  DoDamageChanceInCover: 0
->>>>>>> 351fa8ee12c499a0374be71a192a5f288611f467
-- 
2.49.0


From 8cce356b8e0edda609121fa376809d2e2628ed9d Mon Sep 17 00:00:00 2001
From: Andrey Gumirov <gavt45@bk.ru>
Date: Sat, 7 May 2022 22:56:07 +0700
Subject: [PATCH 15/17] MVP work with ml-agents python library

---
 Assets/Prefabs/Bot.prefab             | 27 +++++++--
 Assets/Prefabs/Player.prefab.meta     |  2 +-
 Assets/Scripts/Character/NPC.cs       | 17 +++++-
 Assets/Settings/Game Settings.asset   |  0
 Packages/manifest.json                |  7 +++
 Packages/packages-lock.json           | 79 +++++++++++++++++++++++++++
 ProjectSettings/ProjectSettings.asset |  6 +-
 7 files changed, 129 insertions(+), 9 deletions(-)
 mode change 100755 => 100644 Assets/Prefabs/Bot.prefab
 mode change 100755 => 100644 Assets/Prefabs/Player.prefab.meta
 mode change 100755 => 100644 Assets/Settings/Game Settings.asset

diff --git a/Assets/Prefabs/Bot.prefab b/Assets/Prefabs/Bot.prefab
old mode 100755
new mode 100644
index 5b177e0..add7a37
--- a/Assets/Prefabs/Bot.prefab
+++ b/Assets/Prefabs/Bot.prefab
@@ -50,6 +50,7 @@ GameObject:
   - component: {fileID: 2756943273076691504}
   - component: {fileID: 6638271233700792696}
   - component: {fileID: 5988018701276467001}
+  - component: {fileID: 1547882613481475944}
   m_Layer: 8
   m_Name: Bot
   m_TagString: Defender
@@ -210,8 +211,8 @@ MonoBehaviour:
   m_Name: 
   m_EditorClassIdentifier: 
   m_SensorName: BufferSensor
-  m_ObservableSize: 0
-  m_MaxNumObservables: 0
+  m_ObservableSize: 5
+  m_MaxNumObservables: 10
 --- !u!114 &2756943273076691504
 MonoBehaviour:
   m_ObjectHideFlags: 0
@@ -225,19 +226,19 @@ MonoBehaviour:
   m_Name: 
   m_EditorClassIdentifier: 
   m_BrainParameters:
-    VectorObservationSize: 1
+    VectorObservationSize: 14
     NumStackedVectorObservations: 1
     m_ActionSpec:
       m_NumContinuousActions: 0
-      BranchSizes: 01000000
-    VectorActionSize: 01000000
+      BranchSizes: 010000000100000001000000
+    VectorActionSize: 010000000100000001000000
     VectorActionDescriptions: []
     VectorActionSpaceType: 0
     hasUpgradedBrainParametersWithActionSpec: 1
   m_Model: {fileID: 0}
   m_InferenceDevice: 0
   m_BehaviorType: 0
-  m_BehaviorName: My Behavior
+  m_BehaviorName: npc
   TeamId: 0
   m_UseChildSensors: 1
   m_UseChildActuators: 1
@@ -272,3 +273,17 @@ MonoBehaviour:
   m_EditorClassIdentifier: 
   navMeshAgent: {fileID: 8774702625908438859}
   flag: {fileID: 6818223691859422291, guid: 1685c1d9ce4ab174f95c646b1826010b, type: 3}
+--- !u!114 &1547882613481475944
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_CorrespondingSourceObject: {fileID: 0}
+  m_PrefabInstance: {fileID: 0}
+  m_PrefabAsset: {fileID: 0}
+  m_GameObject: {fileID: 6171680323948707524}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 3a5c9d521e5ef4759a8246a07d52221e, type: 3}
+  m_Name: 
+  m_EditorClassIdentifier: 
+  DecisionPeriod: 5
+  TakeActionsBetweenDecisions: 0
diff --git a/Assets/Prefabs/Player.prefab.meta b/Assets/Prefabs/Player.prefab.meta
old mode 100755
new mode 100644
index 1fa6659..43cb0ce
--- a/Assets/Prefabs/Player.prefab.meta
+++ b/Assets/Prefabs/Player.prefab.meta
@@ -1,5 +1,5 @@
 fileFormatVersion: 2
-guid: 80f6c1c85e5daed4c96c70205ed5503d
+guid: a7480b77908b042d8adcdd84e8c2c15e
 PrefabImporter:
   externalObjects: {}
   userData: 
diff --git a/Assets/Scripts/Character/NPC.cs b/Assets/Scripts/Character/NPC.cs
index 36b76ec..7af2ae9 100644
--- a/Assets/Scripts/Character/NPC.cs
+++ b/Assets/Scripts/Character/NPC.cs
@@ -73,6 +73,7 @@ public class NPC : Agent, ICharacter
 
     public override void CollectObservations(VectorSensor sensor)
     {
+        // Debug.Log("Collect observations called!");
         navPointIdDict = MapManager.Instance.IDToNavPoint;
         if (navPointIdDict is null)
             Debug.LogError("Cant Find Nav Point Dictionary");
@@ -88,6 +89,7 @@ public class NPC : Agent, ICharacter
         sensor.AddObservation(candidates.Count);
         sensor.AddObservation(moveController.PointStartID);
         sensor.AddObservation(moveController.PointEndID);
+        // Debug.Log("Done common!");
         //state sensors
         sensor.AddObservation((int)NpcState.State);
         sensor.AddObservation((int)NpcBodyState.State);
@@ -95,6 +97,7 @@ public class NPC : Agent, ICharacter
         sensor.AddObservation(navPointIdDict[moveController.PointStartID].DeathAttr);
         sensor.AddObservation(navPointIdDict[moveController.PointEndID].DeathAttr);
         sensor.AddObservation(moveController.FlagDistance);
+        // Debug.Log("Done state sensors!");
 
         //point sensors
         foreach (var point in candidates)
@@ -112,11 +115,14 @@ public class NPC : Agent, ICharacter
                               point.Position).ToInt()
         });
         }
+        // Debug.Log("Done collect observations!");
     }
 
     public override void OnActionReceived(ActionBuffers actions)
     {
+        // Debug.Log("Actions recieved!");
         var result = actions.DiscreteActions;
+        // Debug.Log(result[0] + " " + result[1]);
         if (result[0] == 0)
         {
             if (navPointIdDict[moveController.PointStartID].navType != NavPointType.Cover)
@@ -132,18 +138,26 @@ public class NPC : Agent, ICharacter
                 default: throw new ArgumentException("Undefined Action recieved");
             }
         }
+        // Debug.Log(result[0] == 1);
         if (result[0] == 1)
         {
+            // Debug.Log("BEFORE SOme shitty if >:(");
             if (navPointIdDict[moveController.PointStartID].navType != NavPointType.Direction)
+            {
+                // Debug.Log("SOme shitty if >:(");
                 return;
+            }
+            // Debug.Log("FUCK");
+
             switch (result[1])
             {
                 case 0: moveController.GoToNextNavPoint(navPointIdDict[result[2]]);
-                    NpcState = RunningState; break;
+                    NpcState = RunningState; Debug.Log("Go to point " + result[2]);break;
                 case 1: NpcState = DirectState; break;
                 default: throw new ArgumentException("Undefined Action recieved");
             }
         }
+        
         if (result[0] == 2)
         {
             if (moveController.PointStartID == moveController.PointEndID && moveController.PointEndID != -1)
@@ -155,6 +169,7 @@ public class NPC : Agent, ICharacter
                 default: throw new ArgumentException("Undefined Action recieved");
             }
         }
+        // Debug.Log("Actions processed!");
     }
     #endregion
 
diff --git a/Assets/Settings/Game Settings.asset b/Assets/Settings/Game Settings.asset
old mode 100755
new mode 100644
diff --git a/Packages/manifest.json b/Packages/manifest.json
index a5b072e..e756f82 100755
--- a/Packages/manifest.json
+++ b/Packages/manifest.json
@@ -1,17 +1,24 @@
 {
   "dependencies": {
+    "com.unity.2d.sprite": "1.0.0",
+    "com.unity.2d.tilemap": "1.0.0",
+    "com.unity.ads": "3.7.5",
+    "com.unity.analytics": "3.6.12",
     "com.unity.collab-proxy": "1.14.12",
     "com.unity.ide.rider": "1.2.1",
     "com.unity.ide.visualstudio": "2.0.14",
     "com.unity.ide.vscode": "1.2.4",
     "com.unity.inputsystem": "1.3.0",
     "com.unity.ml-agents": "2.0.1",
+    "com.unity.multiplayer-hlapi": "1.0.8",
     "com.unity.probuilder": "4.5.2",
+    "com.unity.purchasing": "4.1.2",
     "com.unity.test-framework": "1.1.30",
     "com.unity.textmeshpro": "2.1.6",
     "com.unity.timeline": "1.2.18",
     "com.unity.toolchain.win-x86_64-linux-x86_64": "2.0.0",
     "com.unity.ugui": "1.0.0",
+    "com.unity.xr.legacyinputhelpers": "2.1.8",
     "com.unity.modules.ai": "1.0.0",
     "com.unity.modules.androidjni": "1.0.0",
     "com.unity.modules.animation": "1.0.0",
diff --git a/Packages/packages-lock.json b/Packages/packages-lock.json
index bb10cfc..0bac4b2 100755
--- a/Packages/packages-lock.json
+++ b/Packages/packages-lock.json
@@ -1,5 +1,35 @@
 {
   "dependencies": {
+    "com.unity.2d.sprite": {
+      "version": "1.0.0",
+      "depth": 0,
+      "source": "builtin",
+      "dependencies": {}
+    },
+    "com.unity.2d.tilemap": {
+      "version": "1.0.0",
+      "depth": 0,
+      "source": "builtin",
+      "dependencies": {}
+    },
+    "com.unity.ads": {
+      "version": "3.7.5",
+      "depth": 0,
+      "source": "registry",
+      "dependencies": {
+        "com.unity.ugui": "1.0.0"
+      },
+      "url": "https://packages.unity.com"
+    },
+    "com.unity.analytics": {
+      "version": "3.6.12",
+      "depth": 0,
+      "source": "registry",
+      "dependencies": {
+        "com.unity.ugui": "1.0.0"
+      },
+      "url": "https://packages.unity.com"
+    },
     "com.unity.barracuda": {
       "version": "2.0.0",
       "depth": 1,
@@ -86,6 +116,15 @@
       },
       "url": "https://packages.unity.com"
     },
+    "com.unity.multiplayer-hlapi": {
+      "version": "1.0.8",
+      "depth": 0,
+      "source": "registry",
+      "dependencies": {
+        "nuget.mono-cecil": "0.1.6-preview"
+      },
+      "url": "https://packages.unity.com"
+    },
     "com.unity.probuilder": {
       "version": "4.5.2",
       "depth": 0,
@@ -95,6 +134,29 @@
       },
       "url": "https://packages.unity.com"
     },
+    "com.unity.purchasing": {
+      "version": "4.1.2",
+      "depth": 0,
+      "source": "registry",
+      "dependencies": {
+        "com.unity.ugui": "1.0.0",
+        "com.unity.modules.unityanalytics": "1.0.0",
+        "com.unity.modules.unitywebrequest": "1.0.0",
+        "com.unity.modules.jsonserialize": "1.0.0",
+        "com.unity.modules.androidjni": "1.0.0",
+        "com.unity.services.core": "1.0.1"
+      },
+      "url": "https://packages.unity.com"
+    },
+    "com.unity.services.core": {
+      "version": "1.0.1",
+      "depth": 1,
+      "source": "registry",
+      "dependencies": {
+        "com.unity.modules.unitywebrequest": "1.0.0"
+      },
+      "url": "https://packages.unity.com"
+    },
     "com.unity.settings-manager": {
       "version": "1.0.3",
       "depth": 1,
@@ -169,6 +231,23 @@
         "com.unity.modules.imgui": "1.0.0"
       }
     },
+    "com.unity.xr.legacyinputhelpers": {
+      "version": "2.1.8",
+      "depth": 0,
+      "source": "registry",
+      "dependencies": {
+        "com.unity.modules.vr": "1.0.0",
+        "com.unity.modules.xr": "1.0.0"
+      },
+      "url": "https://packages.unity.com"
+    },
+    "nuget.mono-cecil": {
+      "version": "0.1.6-preview",
+      "depth": 1,
+      "source": "registry",
+      "dependencies": {},
+      "url": "https://packages.unity.com"
+    },
     "com.unity.modules.ai": {
       "version": "1.0.0",
       "depth": 0,
diff --git a/ProjectSettings/ProjectSettings.asset b/ProjectSettings/ProjectSettings.asset
index dbf6f6c..b6bb648 100644
--- a/ProjectSettings/ProjectSettings.asset
+++ b/ProjectSettings/ProjectSettings.asset
@@ -614,7 +614,11 @@ PlayerSettings:
   webGLLinkerTarget: 1
   webGLThreadsSupport: 0
   webGLWasmStreaming: 0
-  scriptingDefineSymbols: {}
+  scriptingDefineSymbols:
+    1: CROSS_PLATFORM_INPUT
+    4: CROSS_PLATFORM_INPUT;MOBILE_INPUT
+    7: CROSS_PLATFORM_INPUT;MOBILE_INPUT
+    14: MOBILE_INPUT
   platformArchitecture: {}
   scriptingBackend: {}
   il2cppCompilerConfiguration: {}
-- 
2.49.0


From fa6b5f602ee65d16f94119d8ed74599852c988b4 Mon Sep 17 00:00:00 2001
From: Andrey Gumirov <gavt45@bk.ru>
Date: Sat, 7 May 2022 22:58:25 +0700
Subject: [PATCH 16/17] Added test NB

---
 test-ml-agents.ipynb | 20623 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 20623 insertions(+)
 create mode 100644 test-ml-agents.ipynb

diff --git a/test-ml-agents.ipynb b/test-ml-agents.ipynb
new file mode 100644
index 0000000..1e2d31e
--- /dev/null
+++ b/test-ml-agents.ipynb
@@ -0,0 +1,20623 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "id": "8104e2db-b1a8-40a0-a238-5d9149fd74b0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from mlagents_envs.environment import UnityEnvironment\n",
+    "import mlagents_envs\n",
+    "import numpy as np"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 74,
+   "id": "6f477382-acc9-4aec-907a-7f58caf955ed",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import random"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 83,
+   "id": "b7f60f26-0a90-4ea5-b2c2-b5683bda56a6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "env = UnityEnvironment()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 84,
+   "id": "5929b410-12c3-4bd9-b984-b2c29a76c3f3",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "env.reset()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 85,
+   "id": "f108ff09-9f42-4405-add3-6df941c48f8b",
+   "metadata": {
+    "scrolled": true,
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052cafd0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0220>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0af0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb880>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca09a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb5b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebb20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebc40>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0af0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0b50>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca07f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107434eb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0460>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107434eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0eb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0190>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052bbd90>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cafd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052bbd90>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca100>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0f70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca07f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0ca0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 78.21462], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0af0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca08e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0af0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67fa0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  1.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebeb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0220>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052caf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb100>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebeb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0b50>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052bbd90>, <mlagents_envs.base_env.TerminalSteps object at 0x1052caf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0f70>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0be0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  1.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb5b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb5b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebfd0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb7f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb5b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052caf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce07c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebaf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb700>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebee0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cad60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  1.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 80.694435], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb100>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebb20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebdf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebeb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb6d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb760>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x10744b7f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb940>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb0d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebeb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb6d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceba30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x1052caf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebdf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb100>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb7f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67340>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67fa0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0eb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0be0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0c70>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0220>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca07f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0190>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0220>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0fd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0eb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x10744b7f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca08e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0220>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.29597], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebc40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0ca0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0b50>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67fa0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebc40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0220>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb430>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb100>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca07f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb6d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebc40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb430>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb6d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0fd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  1.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052cafd0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca07f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0220>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67fa0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca09a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0220>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca08e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  1.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca07f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebaf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0f70>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb760>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 90.925804], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb5b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb7f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca08e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb5b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb760>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0fd0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0b50>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca08e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0460>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0190>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca08e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0220>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0f70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0250>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce07c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0190>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0af0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0eb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67fa0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.28178], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0250>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0f70>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052caf40>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cafd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0c70>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca07f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce07c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0c70>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca100>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce07c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce07c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0fd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0730>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0220>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0eb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67fa0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0af0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0b50>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca07f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052caf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0b50>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0be0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26389], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0190>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca08e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebeb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebeb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67fa0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cad60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0fd0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67fa0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0520>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca08e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca08e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca00a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0be0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x10744b7f0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052caf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0ca0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x10744b7f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x10744b7f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca00a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca08e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0220>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0be0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0790>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052cad60>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0220>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x1052bbd90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0ca0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca07f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebc40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd30>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebb20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0fd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceba00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052bbd90>, <mlagents_envs.base_env.TerminalSteps object at 0x1052caf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0220>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0f70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0b50>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca08e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb490>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0af0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0be0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca08e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0f70>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0b50>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0be0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67340>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb880>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebb20>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb790>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb0d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceba00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb100>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce07c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce07c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67fa0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca00a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb490>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceba00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0eb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 83.26209], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb940>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb6d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb880>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceba30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb0d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebb20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebdf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0be0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x10744b7f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca09a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce07c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0220>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca100>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cafd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0fd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107434eb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052bbd90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x1052caf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0c40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca07f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceba00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca100>, <mlagents_envs.base_env.TerminalSteps object at 0x1052caf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0ca0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0190>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0220>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb430>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0b50>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca100>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebee0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0250>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0190>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052cafd0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb0d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceba30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb6d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0fd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0220>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb880>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb6d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67fa0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0ca0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0be0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67fa0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0220>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0220>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca100>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb100>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0be0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca00a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0460>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebc40>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebeb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0790>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca100>, <mlagents_envs.base_env.TerminalSteps object at 0x1052caf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0b50>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0f70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0730>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67340>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0af0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0790>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0eb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebeb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebb20>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25988], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceba00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb6d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb8e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebb20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0790>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca00a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca00a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0c40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0ca0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca08e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0490>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb490>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb7f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb880>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb760>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca07f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebeb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67fa0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb700>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebee0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb5b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebc40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb490>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebb20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb700>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebc40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26341], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebeb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebb20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebeb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb0d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67fa0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb880>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0220>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb6d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebc40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb880>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb6d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca09a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb6d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb5b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x10744b7f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107434eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca07f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca08e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0220>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0b50>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce07c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052bbd90>, <mlagents_envs.base_env.TerminalSteps object at 0x107434eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca07f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30755], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107434eb0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca08e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67fa0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0460>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052cafd0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca100>, <mlagents_envs.base_env.TerminalSteps object at 0x1052caf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce07c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca07f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb8e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0460>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb5b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67fa0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebc40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0490>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0730>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0730>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052caf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce07c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0ca0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce07c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0460>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0250>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce07c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0190>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67fa0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 85.95022], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0250>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67fa0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052cafd0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cad60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052caf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0be0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0190>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca00a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67fa0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0730>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67fa0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0790>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052caf40>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cafd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67fa0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67340>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0f70>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0c40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca08e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb100>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb100>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca09a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0af0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebeb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb100>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb6d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb520>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebee0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebfd0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.14938], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebee0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb430>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb0d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67fa0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0c40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb6d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebc40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0220>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca07f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceba00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0f70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb760>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb5b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb490>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0460>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebaf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb5b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb0d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb430>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb100>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb790>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebdf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.25333], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb490>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebc40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebfd0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebee0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebdf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebeb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebb20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb0d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb100>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebdf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebb20>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca07f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0ca0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca09a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0b50>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0af0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0b50>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0fd0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107434eb0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cafd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb430>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebdf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb760>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebeb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebfd0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb760>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceba00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  1.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb100>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb5b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebeb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb490>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb7f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb8e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb0d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0c40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca09a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb5b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  1.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceba00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb0d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb5b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb760>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca100>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  1.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb880>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        1.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb100>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebee0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  0.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        0.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebb20>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  2.      ,  2.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.      ,  0.      ,  0.      ,  3.      ,  3.      ,  3.      ,\n",
+      "        2.      ,  1.      ,  1.      ,  1.      ,  0.      ,  0.      ,\n",
+      "        0.      , 91.273026], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0520>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb0d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0460>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceba30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebdf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb880>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebb20>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb8e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb490>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb940>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb490>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb0d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca08e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0fd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebaf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052caf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0190>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0fd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca07f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0730>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0b50>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0b50>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d00>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67a90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0220>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.27214], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca08e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0790>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0ca0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb520>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb0d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb0d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebee0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c672e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb940>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb760>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb880>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebb20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67e80>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb760>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebdf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb940>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebeb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebaf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb0d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebc40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebee0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0b50>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107434eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0250>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67340>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0730>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb7f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebeb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebfd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca08e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d90>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0730>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        2.    ,  1.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0af0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0250>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca08e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0ca0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb8e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb7f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebdf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052ca9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052caf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebeb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb760>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb8e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        1.    ,  1.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb880>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb7f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebeb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb0d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        2.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052caf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  2.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107434eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  1.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebaf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebeb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        0.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.    ,  0.    ,  0.    ,  3.    ,  3.    ,  3.    ,  0.    ,\n",
+      "        1.    ,  2.    ,  1.    ,  0.    ,  0.    ,  0.    , 91.2745],\n",
+      "      dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce07c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0460>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0490>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0c40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca08e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca00a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0c40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0190>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb490>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb760>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebeb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb8e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebeb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb490>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb0d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb790>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebb20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebdf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb430>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb8e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebdf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebaf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0970>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0160>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb8e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebfd0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb0d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb5b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cad60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb760>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebaf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb0d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb8e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67fa0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb0d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebb20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb5b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb5b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb7f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0520>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebb20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67340>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebb20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce07c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce04c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce02b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce07c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052bbd90>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce0520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052cafd0>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cad60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cafd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.26514], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca02e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x10744b7f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107434eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0ca0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca00a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0d60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce02b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca02e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0520>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67a90>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca100>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67fa0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca08e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0730>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0a30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0190>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0c40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0be0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0b20>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0f70>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cad60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce04c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ce0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0eb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107ce01f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0790>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0490>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca01f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0c40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052caf40>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x1052cafd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052cad60>, <mlagents_envs.base_env.TerminalSteps object at 0x1052ca100>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107c674c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0ca0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052bbd90>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0490>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0610>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb430>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb9d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3a0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebfd0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb5b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb4f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebdf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0eb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb100>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0250>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0190>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0c40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd00d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0d00>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0610>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0d30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0fd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb100>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceba30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c674c0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb7f0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0ac0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb6d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb490>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb790>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0370>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebaf0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0d60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x1052bbd90>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb520>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb4f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67370>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb430>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebaf0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebb20>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb700>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb940>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67e80>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb8e0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebee0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67bb0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb790>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb5b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0970>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67fa0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb9d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb100>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0070>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb3d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb0d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb5b0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebee0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebf40>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb880>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0b20>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67bb0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebee0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd03d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebf40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb520>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ca0af0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0a60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107ca0fd0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd60>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd30>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67910>, <mlagents_envs.base_env.TerminalSteps object at 0x107c672e0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb880>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3a0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0160>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67340>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67370>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebb20>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebd60>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107c67340>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd01c0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb430>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb7f0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb070>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb5b0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  1.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd00d0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0340>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0ac0>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd0580>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        2.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb100>, <mlagents_envs.base_env.TerminalSteps object at 0x107cebc40>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cebd30>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb3d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  1.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0580>, <mlagents_envs.base_env.TerminalSteps object at 0x107c67910>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        1.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107ceb700>, <mlagents_envs.base_env.TerminalSteps object at 0x107ceb0d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  0.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n",
+      "Step\n",
+      "(<mlagents_envs.base_env.DecisionSteps object at 0x107cd0940>, <mlagents_envs.base_env.TerminalSteps object at 0x107cd03d0>)\n",
+      "DecisionStep(obs=[array([[0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 0., 0., 0.],\n",
+      "       [0., 1., 1., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.],\n",
+      "       [0., 0., 0., 0., 0.]], dtype=float32), array([ 0.     ,  0.     ,  0.     ,  3.     ,  3.     ,  3.     ,\n",
+      "        0.     ,  2.     ,  2.     ,  1.     ,  0.     ,  0.     ,\n",
+      "        0.     , 91.30669], dtype=float32)], reward=0.0, agent_id=0, action_mask=[array([False]), array([False]), array([False])], group_id=0, group_reward=0.0)\n",
+      "0.0\n"
+     ]
+    },
+    {
+     "ename": "UnityCommunicatorStoppedException",
+     "evalue": "Communicator has exited.",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mUnityCommunicatorStoppedException\u001b[0m         Traceback (most recent call last)",
+      "Input \u001b[0;32mIn [85]\u001b[0m, in \u001b[0;36m<cell line: 2>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mwhile\u001b[39;00m \u001b[38;5;28;01mTrue\u001b[39;00m:\n\u001b[0;32m----> 2\u001b[0m     \u001b[43menv\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mstep\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m      3\u001b[0m     \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mStep\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m      4\u001b[0m     asd \u001b[38;5;241m=\u001b[39m env\u001b[38;5;241m.\u001b[39mget_steps(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mnpc?team=0\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
+      "File \u001b[0;32m~/opt/miniforge3/lib/python3.9/site-packages/mlagents_envs/timers.py:305\u001b[0m, in \u001b[0;36mtimed.<locals>.wrapped\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    303\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mwrapped\u001b[39m(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs):\n\u001b[1;32m    304\u001b[0m     \u001b[38;5;28;01mwith\u001b[39;00m hierarchical_timer(func\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__qualname__\u001b[39m):\n\u001b[0;32m--> 305\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfunc\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/opt/miniforge3/lib/python3.9/site-packages/mlagents_envs/environment.py:350\u001b[0m, in \u001b[0;36mUnityEnvironment.step\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    348\u001b[0m     outputs \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_communicator\u001b[38;5;241m.\u001b[39mexchange(step_input, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_poll_process)\n\u001b[1;32m    349\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m outputs \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m--> 350\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m UnityCommunicatorStoppedException(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mCommunicator has exited.\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m    351\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_update_behavior_specs(outputs)\n\u001b[1;32m    352\u001b[0m rl_output \u001b[38;5;241m=\u001b[39m outputs\u001b[38;5;241m.\u001b[39mrl_output\n",
+      "\u001b[0;31mUnityCommunicatorStoppedException\u001b[0m: Communicator has exited."
+     ]
+    }
+   ],
+   "source": [
+    "while True:\n",
+    "    env.step()\n",
+    "    print(\"Step\")\n",
+    "    asd = env.get_steps('npc?team=0')\n",
+    "    print(asd)\n",
+    "    print(asd[0][0])\n",
+    "    _id = asd[0][0].obs[0][0][0]\n",
+    "    print(_id)\n",
+    "    env.set_action_for_agent('npc?team=0', 0, mlagents_envs.environment.ActionTuple(discrete=np.array([[1, 0, random.randint(0,2)]])))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 86,
+   "id": "db100c84-22ab-491b-b68d-4d5c1bbc66a3",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "env.close()"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
-- 
2.49.0


From aa7b50554b7b572df14bc2a184b240e4a972b78e Mon Sep 17 00:00:00 2001
From: Andrey Gumirov <gavt45@bk.ru>
Date: Sat, 7 May 2022 23:00:24 +0700
Subject: [PATCH 17/17] Added project version

---
 ProjectSettings/ProjectVersion.txt | 2 ++
 1 file changed, 2 insertions(+)
 create mode 100644 ProjectSettings/ProjectVersion.txt

diff --git a/ProjectSettings/ProjectVersion.txt b/ProjectSettings/ProjectVersion.txt
new file mode 100644
index 0000000..bf8f7eb
--- /dev/null
+++ b/ProjectSettings/ProjectVersion.txt
@@ -0,0 +1,2 @@
+m_EditorVersion: 2019.4.36f1
+m_EditorVersionWithRevision: 2019.4.36f1 (660c164b2fc5)
-- 
2.49.0