def play_game(players,
num_samples=100,
num_simulations=None,
total_simulation_seconds=1):
tiles_by_player = deal_tiles()
first_player = random.choice([0, 1, 2, 3])
state = DominoState(first_player, {
'tiles_by_player': tiles_by_player,
'suits_at_ends': set()
})
game = [state]
log(f"Starts player {first_player}")
log("Tiles : ")
log(pformat(state._tiles_by_player))
while not state.is_terminal():
log("=======================================")
log(pformat(state._tiles_by_player[state._current_player]))
state = play_with_algo(
players[state._current_player],
state,
game,
num_samples,
num_simulations=num_simulations,
total_simulation_seconds=total_simulation_seconds)
game.append(state)
print_state(state)
log(f"winneeeer {state.calc_reward()}")
log(pformat(state._tiles_by_player))
record_winner(state._tiles_by_player)
return (game, state.calc_reward())
def test_calc_reward_team2_win_by_points(self):
state = DominoState( 1,{
'tiles_by_player':[
[{5,4}],
[{3, 5}],
[{6} ],
[{5}]
],
'suits_at_ends':{2}
})
self.assertEqual(state.calc_reward(), -1)
def test_calc_reward_team1_win(self):
state = DominoState( 1,{
'tiles_by_player':[
[],
[{3, 5}],
[{4} ],
[{5}]
],
'suits_at_ends':{4,3}
})
self.assertEqual(state.calc_reward(), 1)
class GameOfDomino(TwoPlayersGame):
def __init__(self, players):
self.players = players
self.nplayer = 1 # player 1 starts
self._state = DominoState()
def possible_moves(self):
return self._state.get_possible_actions()
def make_move(self, move):
# self.pile -= int(move) # remove bones.
self._state = self._state.next_state_from_action(move)
def win(self):
# return self.pile<=0 # opponent took the last bone ?
winner = self._state.calc_reward()
team_1 = self._state.team_1
team_2 = self._state.team_2
if winner == 0:
return False
if winner == team_1 and self.nplayer == 1:
return True
if winner == team_2 and self.nplayer == 2:
return True
return False
def is_over(self):
# return self.win() # Game stops when someone wins.
return self._state.is_terminal()
def show(self):
# print ("%d bones left in the pile" % self.pile)
print(self._state)
def scoring(self):
return 100 if self.win() else 0 # For the AI