def main():
mcts = MCTS()
game = Othello()
for i in range(32):
action = mcts.get_action(game)
game.move(action)
if game.game_over():
break
actions = game.get_actions()
probs = np.ones(actions.shape[0])
action = sample(probs, actions)
game.move(action)
if game.game_over():
break
# input('waiting')
print(game.get_true_state())
print(game.get_score())
print(game.get_winner())
print('game over')
class ReversiEnv(object):
"""docstring for ReversiEnv."""
def __init__(self):
super(ReversiEnv, self).__init__()
self.n = 8
def length(self):
return self.n * self.n
def reset(self):
self.game = Othello()
return self.game.get_state(), self.game.get_turn()
def action_space(self):
return self.game.get_actions()
def step(self, action):
self.game.move(action)
reward = self.game.get_winner()
return self.game.get_state(), self.game.get_turn(), reward