From b699258e769429f881fdef34ca947e1311983be3 Mon Sep 17 00:00:00 2001 From: rtz19970824 Date: Thu, 28 Dec 2017 15:55:07 +0800 Subject: [PATCH] debug for reversi --- AlphaGo/game.py | 2 +- AlphaGo/player.py | 7 +++---- AlphaGo/reversi.py | 2 +- 3 files changed, 5 insertions(+), 6 deletions(-) diff --git a/AlphaGo/game.py b/AlphaGo/game.py index f17c7af..82cf254 100644 --- a/AlphaGo/game.py +++ b/AlphaGo/game.py @@ -123,7 +123,7 @@ class Game: sys.stdout.flush() if __name__ == "__main__": - game = Game(name="go", checkpoint_path="./checkpoint") + game = Game(name="reversi", checkpoint_path=None) game.debug = True game.think_play_move(utils.BLACK) diff --git a/AlphaGo/player.py b/AlphaGo/player.py index 8d46ae5..bd2a2d1 100644 --- a/AlphaGo/player.py +++ b/AlphaGo/player.py @@ -18,12 +18,11 @@ class Player(object): if __name__ == '__main__': parser = argparse.ArgumentParser() - parser.add_argument("--checkpoint_path", type=str, default=None) + parser.add_argument("--checkpoint_path", type=str, default="None") parser.add_argument("--role", type=str, default="unknown") - parser.add_argument("--debug", type=str, default=False) - parser.add_argument("--game", type=str, default=False) + parser.add_argument("--debug", type=str, default="False") + parser.add_argument("--game", type=str, default="go") args = parser.parse_args() - if args.checkpoint_path == 'None': args.checkpoint_path = None game = Game(name=args.game, role=args.role, diff --git a/AlphaGo/reversi.py b/AlphaGo/reversi.py index 08a5ec5..1685b66 100644 --- a/AlphaGo/reversi.py +++ b/AlphaGo/reversi.py @@ -99,7 +99,7 @@ class Reversi: def simulate_hashable_conversion(self, state): # since go is MDP, we only need the last board for hashing - return tuple(state[0][-1]) + return tuple(state[0][-1].flatten().tolist()) def _get_winner(self, board): black_num, white_num = self._number_of_black_and_white(board)