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
This commit is contained in:
@@ -1,7 +1,7 @@
|
|||||||
- itemType: ACTIONS
|
- itemType: ACTIONS
|
||||||
type: NODE
|
type: NODE
|
||||||
- itemType: STEPS
|
- itemType: STEPS
|
||||||
steps: 13
|
steps: 15
|
||||||
- itemType: PORTS
|
- itemType: PORTS
|
||||||
portsList:
|
portsList:
|
||||||
- port: '21'
|
- port: '21'
|
||||||
@@ -42,7 +42,7 @@
|
|||||||
- itemType: RED_POL
|
- itemType: RED_POL
|
||||||
id: '2'
|
id: '2'
|
||||||
startStep: 3
|
startStep: 3
|
||||||
endStep: 13
|
endStep: 15
|
||||||
targetNodeId: '1'
|
targetNodeId: '1'
|
||||||
initiator: DIRECT
|
initiator: DIRECT
|
||||||
type: FILE
|
type: FILE
|
||||||
@@ -66,7 +66,7 @@
|
|||||||
- itemType: RED_POL
|
- itemType: RED_POL
|
||||||
id: '4'
|
id: '4'
|
||||||
startStep: 6
|
startStep: 6
|
||||||
endStep: 13
|
endStep: 15
|
||||||
targetNodeId: '1'
|
targetNodeId: '1'
|
||||||
initiator: DIRECT
|
initiator: DIRECT
|
||||||
type: OPERATING
|
type: OPERATING
|
||||||
@@ -90,7 +90,7 @@
|
|||||||
- itemType: RED_POL
|
- itemType: RED_POL
|
||||||
id: '6'
|
id: '6'
|
||||||
startStep: 9
|
startStep: 9
|
||||||
endStep: 13
|
endStep: 15
|
||||||
targetNodeId: '1'
|
targetNodeId: '1'
|
||||||
initiator: DIRECT
|
initiator: DIRECT
|
||||||
type: SERVICE
|
type: SERVICE
|
||||||
@@ -114,7 +114,7 @@
|
|||||||
- itemType: RED_POL
|
- itemType: RED_POL
|
||||||
id: '8'
|
id: '8'
|
||||||
startStep: 12
|
startStep: 12
|
||||||
endStep: 13
|
endStep: 15
|
||||||
targetNodeId: '1'
|
targetNodeId: '1'
|
||||||
initiator: DIRECT
|
initiator: DIRECT
|
||||||
type: OS
|
type: OS
|
||||||
|
|||||||
@@ -28,4 +28,4 @@ def test_rewards_are_being_penalised_at_each_step_function():
|
|||||||
Average Reward: 2 (26 / 13)
|
Average Reward: 2 (26 / 13)
|
||||||
"""
|
"""
|
||||||
print("average reward", env.average_reward)
|
print("average reward", env.average_reward)
|
||||||
assert env.average_reward == 2.0
|
assert env.average_reward == -8.0
|
||||||
|
|||||||
Reference in New Issue
Block a user