1443 - updated test_reward.py to reflect updates to reward.py so that the correct config values are called i.e. compromisedShouldBeGood on the correct steps during the training run

tests/config/one_node_states_on_off_lay_down_config.yaml

@@ -1,7 +1,7 @@
 - itemType: ACTIONS
   type: NODE
 - itemType: STEPS
-  steps: 13
+  steps: 15
 - itemType: PORTS
   portsList:
   - port: '21'
@@ -42,7 +42,7 @@
 - itemType: RED_POL
   id: '2'
   startStep: 3
-  endStep: 13
+  endStep: 15
   targetNodeId: '1'
   initiator: DIRECT
   type: FILE
@@ -66,7 +66,7 @@
 - itemType: RED_POL
   id: '4'
   startStep: 6
-  endStep: 13
+  endStep: 15
   targetNodeId: '1'
   initiator: DIRECT
   type: OPERATING
@@ -90,7 +90,7 @@
 - itemType: RED_POL
   id: '6'
   startStep: 9
-  endStep: 13
+  endStep: 15
   targetNodeId: '1'
   initiator: DIRECT
   type: SERVICE
@@ -114,7 +114,7 @@
 - itemType: RED_POL
   id: '8'
   startStep: 12
-  endStep: 13
+  endStep: 15
   targetNodeId: '1'
   initiator: DIRECT
   type: OS

tests/test_reward.py

@@ -28,4 +28,4 @@ def test_rewards_are_being_penalised_at_each_step_function():
         Average Reward: 2 (26 / 13)
     """
     print("average reward", env.average_reward)
-    assert env.average_reward == 2.0
+    assert env.average_reward == -8.0