def getReward(self, state, action, transitionState):
isDone, isFirst, isDoubleDown, hasAce, hardCount, dealerSoftCount = transitionState
multiplier = 2 if isDoubleDown else 1
if hardCount > 21:
return -2 * multiplier
softCount = hardCount + 10 if hasAce and hardCount <= 11 else hardCount
if isDone:
if isFirst and softCount == 21:
return multiplier
# Simulate the dealer's actions
dealerAgent = DealerAgent()
dealerCardValue = dealerSoftCount - 1 if dealerSoftCount != 11 else 0
card = Card(0, dealerCardValue)
dealerHand = Hand(1)
dealerHand.addCard(card)
deck = Deck(1, 4, 13)
dealerHand.addCard(deck.take())
while dealerAgent.getNextAction(None, dealerHand) == Actions.HIT:
dealerHand.addCard(deck.take())
return (
multiplier
if softCount > dealerHand.getSoftCount()
else (0 if softCount == dealerHand.getSoftCount() else -multiplier)
)
else:
return 0
def test_4CardHand(self):
hand = Hand(0)
hand.addCard(Card(1, 6))
hand.addCard(Card(1, 4))
hand.addCard(Card(1, 2))
hand.addCard(Card(1, 0))
dealer = DealerAgent()
self.assertEqual(Actions.HIT, dealer.getNextAction(None, hand))
def getReward(self, state, action, transitionState):
isDone, isFirst, isDoubleDown, hasAce, hardCount, dealerSoftCount = transitionState
multiplier = 2 if isDoubleDown else 1
if hardCount > 21:
return -2 * multiplier
softCount = hardCount + 10 if hasAce and hardCount <= 11 else hardCount
if isDone:
if isFirst and softCount == 21:
return multiplier
# Simulate the dealer's actions
dealerAgent = DealerAgent()
dealerCardValue = dealerSoftCount - 1 if dealerSoftCount != 11 else 0
card = Card(0, dealerCardValue)
dealerHand = Hand(1)
dealerHand.addCard(card)
deck = Deck(1, 4, 13)
dealerHand.addCard(deck.take())
while dealerAgent.getNextAction(None, dealerHand) == Actions.HIT:
dealerHand.addCard(deck.take())
return multiplier if softCount > dealerHand.getSoftCount() else (
0 if softCount == dealerHand.getSoftCount() else -multiplier)
else:
return 0
elif playerAgentString == "StandingAgent":
playerAgents.append(StandingAgent())
elif playerAgentString == "NoBustAgent":
playerAgents.append(NoBustAgent())
elif playerAgentString == "HumanAgent":
playerAgents.append(HumanAgent())
elif playerAgentString == "ValueIterationAgent":
if not iterations:
print "Number of iterations must be specified with ValueIterationAgent"
sys.exit(1)
playerAgents.append(ValueIterationAgent(iterations))
else:
print "Unrecognized agent {0}".format(playerAgentString)
sys.exit(1)
dealerAgent = DealerAgent()
if trainingRounds > 0:
trainingAgents = filter(lambda x: x.needsTraining(), playerAgents)
print "Training ({0} rounds)...".format(trainingRounds)
game = Game(dealerAgent, trainingAgents)
game.executeGame(trainingRounds)
for playerAgent in playerAgents:
playerAgent.trainingOver()
print "Testing ({0} rounds)...".format(realRounds)
game = Game(dealerAgent, playerAgents)
game.executeGame(realRounds)
print game.resultString()
for (playerAgentString, playerAgent) in zip(playerAgentStrings, playerAgents):
def test_Soft17(self):
hand = Hand(0)
hand.addCard(Card(1, 6))
hand.addCard(Card(1, 0))
dealer = DealerAgent()
self.assertEqual(Actions.STAND, dealer.getNextAction(None, hand))
