def play_against_minimax(self, color, game_count):
"""
lets the agent play against an optimal minimax player. since the minimax player
has an optimal strategy the best score the agent can get is 0.5
:param color: the color of the agent
:param game_count: the number of games that are played
:return: the mean score against the minimax player 0: lose, 0.5 draw, 1: win
"""
q_player = tournament.QNetPlayer(self.network)
minimax_player = tournament.MinimaxPlayer()
score = tournament.play_one_color(game_count, q_player, color, minimax_player)
return score
def net_vs_minimax(net, game_count, mcts_sim_count, c_puct, temp, color=None):
"""
lets the alpha zero network play against a minimax player
:param net: alpha zero network
:param game_count: total games to play
:param mcts_sim_count number of monte carlo simulations
:param c_puct constant that controls the exploration
:param temp the temperature
:param color the color of the network
:return: score of network
"""
az_player = tournament.AlphaZeroPlayer(net, c_puct, mcts_sim_count, temp)
minimax_player = tournament.MinimaxPlayer()
if color is None:
az_score = tournament.play_match(game_count, az_player, minimax_player)
else:
az_score = tournament.play_one_color(game_count, az_player, color, minimax_player)
return az_score
# play random vs random
game_count = 1000
player1 = tournament.RandomPlayer()
player2 = tournament.RandomPlayer()
white_score = tournament.play_one_color(game_count, player1, CONST.WHITE,
player2)
logger.info("white score for random vs random: {}".format(white_score))
black_score = tournament.play_one_color(game_count, player1, CONST.BLACK,
player2)
logger.info("black score for random vs random: {}".format(black_score))
# play minimax vs minimax to check if the score is 0.5
game_count = 100
player1 = tournament.MinimaxPlayer()
player2 = tournament.MinimaxPlayer()
player1_score = tournament.play_match(game_count, player1, player2)
logger.info("minimax vs minimax score: {}".format(player1_score))
# play random vs minimax
game_count = 1000
player1 = tournament.MinimaxPlayer()
player2 = tournament.RandomPlayer()
white_score = tournament.play_one_color(game_count, player1, CONST.WHITE,
player2)
logger.info("white score for minimax vs random: {}".format(white_score))
black_score = tournament.play_one_color(game_count, player1, CONST.BLACK,
player2)
logger.info("black score for minimax vs random: {}".format(black_score))