fix hurry reward/ add awful first training

Project/Assets/ML-Agents/Examples/SharedAssets/Scripts/RewardManager.cs

     public void UpdateReward(string key, float rawVal)
     {
         float val = rawVal * rewardsDict[key].rewardScalar;
-        rewardsDict[key].maxRewardThisSession =1/maxSteps;
+        rewardsDict[key].maxRewardThisSession = val * maxSteps;
         rewardsDict[key].rewardThisStep = val;
         rewardsDict[key].cumulativeThisEpisode += val;
         rewardsDict[key].cumulativeThisSession += val;

Project/Assets/ML-Agents/Examples/Walker/Scenes/WalkerDynamic.unity

       propertyPath: m_LocalEulerAnglesHint.z
       value: 0
       objectReference: {fileID: 0}
+    - target: {fileID: 6065910098925129095, guid: f51e8260728fd4c8fa87bcda9d0e2027,
+        type: 3}
+      propertyPath: m_Model
+      value: 
+      objectReference: {fileID: 11400000, guid: c85becd32a14a4c048895b4eafc718ce,
+        type: 3}
     m_RemovedComponents: []
   m_SourcePrefab: {fileID: 100100000, guid: f51e8260728fd4c8fa87bcda9d0e2027, type: 3}
 --- !u!4 &6065910099080495282 stripped

Project/Assets/ML-Agents/Examples/Walker/Scripts/WalkerAgent.cs

         rewardManager.UpdateReward("matchSpeed", matchSpeedReward);
         rewardManager.UpdateReward("lookAtTarget", lookAtTargetReward);
         rewardManager.UpdateReward("headHeightOverFeet", headHeightOverFeetReward);
-        rewardManager.UpdateReward("hurryUp", hurryUpReward); 
+        rewardManager.UpdateReward("hurryUp", hurryUpReward/MaxStep); 
     }
     
 //    void FixedUpdate()

Project/Assets/ML-Agents/Examples/Walker/TFModels/WalkerDynamic 1.nn.meta

+fileFormatVersion: 2
+guid: c85becd32a14a4c048895b4eafc718ce
+ScriptedImporter:
+  fileIDToRecycleName:
+    11400000: main obj
+    11400002: model data
+  externalObjects: {}
+  userData: 
+  assetBundleName: 
+  assetBundleVariant: 
+  script: {fileID: 11500000, guid: 19ed1486aa27d4903b34839f37b8f69f, type: 3}