def findBestNoteUCB(self, node: Node):
parentVisit = node.getAction().getVisitCount()
bestchildren = []
bestScore = 0.0
for c in node.getChilds(): #type; State
score = self.ucbValue(parentVisit, c.getAction().getWinScore(), c.getAction().getVisitCount())
if score == bestScore:
bestchildren.append(c)
if score > bestScore:
bestchildren = [c]
bestScore = score
if len(bestchildren) == 0:
logger.warning("No best Children Found")
return random.choice(bestchildren)
def simulateRound(self, node: Node):
rnd = get_round_from_player_round(node.getAction().getRound(),
node.getAction().getRound().hands)
rnd.action_play_card(node.getAction().getCard())
cards = rnd.nr_played_cards
randomPlayer = RandomPlayerSchieber()
while cards < 36:
player_rnd = PlayerRoundCheating()
player_rnd.set_from_round(rnd)
card_action = randomPlayer.play_card(player_rnd)
rnd.action_play_card(card_action)
cards += 1
myPoints = rnd.points_team_0
pointsEnemy = rnd.points_team_1
maxPoints = myPoints + pointsEnemy
if myPoints > pointsEnemy:
return (myPoints - 0) / (maxPoints - 0)
else:
return 0
def expandNode(self, node: Node, rnd: PlayerRoundCheating):
validCards = np.flatnonzero(rnd.get_valid_cards())
for card in validCards:
newNode = Node()
newNode.setParent(node)
newNode.getAction().setRound(rnd)
newNode.getAction().setPlayerNr(node.getAction().getRound().player)
newNode.getAction().setCard(card)
node.addChild(newNode)
def __init__(self) -> None:
self._rootNode = Node()