@@ -487,10 +487,8 @@ public void TestCumulativeReward()
487
487
agent1 . LazyInitialize ( ) ;
488
488
agent2 . SetPolicy ( new TestPolicy ( ) ) ;
489
489
490
- int expectedAgent1Resets = 0 ;
491
- int expectedAgent1ActionSinceReset = 0 ;
490
+ var expectedAgent1ActionSinceReset = 0 ;
492
491
493
- var j = 0 ;
494
492
for ( var i = 0 ; i < 50 ; i ++ )
495
493
{
496
494
expectedAgent1ActionSinceReset += 1 ;
@@ -518,30 +516,46 @@ public void TestMaxStepsReset()
518
516
decisionRequester . DecisionPeriod = 1 ;
519
517
decisionRequester . Awake ( ) ;
520
518
521
- var maxStep = 6 ;
519
+ const int maxStep = 6 ;
522
520
agent1 . maxStep = maxStep ;
523
521
agent1 . LazyInitialize ( ) ;
524
522
525
- int expectedResets = 0 ;
526
- int expectedAgentAction = 0 ;
527
- int expectedAgentActionSinceReset = 0 ;
523
+ var expectedAgentStepCount = 0 ;
524
+ var expectedResets = 0 ;
525
+ var expectedAgentAction = 0 ;
526
+ var expectedAgentActionSinceReset = 0 ;
527
+ var expectedCollectObsCalls = 0 ;
528
+ var expectedCollectObsCallsSinceReset = 0 ;
528
529
529
530
for ( var i = 0 ; i < 15 ; i ++ )
530
531
{
532
+ // Agent should observe and act on each Academy step
531
533
expectedAgentAction += 1 ;
532
534
expectedAgentActionSinceReset += 1 ;
533
- if ( expectedAgentActionSinceReset == maxStep || ( i == 0 ) ) {
535
+ expectedCollectObsCalls += 1 ;
536
+ expectedCollectObsCallsSinceReset += 1 ;
537
+ expectedAgentStepCount += 1 ;
538
+
539
+ // If the next step will put the agent at maxSteps, we expect it to reset
540
+ if ( agent1 . GetStepCount ( ) == maxStep - 1 || ( i == 0 ) )
541
+ {
534
542
expectedResets += 1 ;
535
543
}
536
544
537
- if ( expectedAgentActionSinceReset == maxStep ) {
545
+ if ( agent1 . GetStepCount ( ) == maxStep - 1 )
546
+ {
538
547
expectedAgentActionSinceReset = 0 ;
548
+ expectedCollectObsCallsSinceReset = 0 ;
549
+ expectedAgentStepCount = 0 ;
539
550
}
540
551
aca . EnvironmentStep ( ) ;
541
552
542
- Assert . AreEqual ( expectedAgentAction , agent1 . agentActionCalls ) ;
553
+ Assert . AreEqual ( expectedAgentStepCount , agent1 . GetStepCount ( ) ) ;
543
554
Assert . AreEqual ( expectedResets , agent1 . agentResetCalls ) ;
555
+ Assert . AreEqual ( expectedAgentAction , agent1 . agentActionCalls ) ;
544
556
Assert . AreEqual ( expectedAgentActionSinceReset , agent1 . agentActionCallsSinceLastReset ) ;
557
+ Assert . AreEqual ( expectedCollectObsCalls , agent1 . collectObservationsCalls ) ;
558
+ Assert . AreEqual ( expectedCollectObsCallsSinceReset , agent1 . collectObservationsCallsSinceLastReset ) ;
545
559
}
546
560
}
547
561
}
0 commit comments