def test(self, episodes=100, draw=False):
players = [TDAgent(Game.TOKENS[0], self), RandomAgent(Game.TOKENS[1])]
winners = [0, 0]
for episode in range(episodes):
game = Game.new()
winner = game.play(players, draw=draw)
winners[winner] += 1
winners_total = sum(winners)
print("[Episode %d] %s (%s) vs %s (%s) %d:%d of %d games (%.2f%%)" % (episode, \
players[0].name, players[0].player, \
players[1].name, players[1].player, \
winners[0], winners[1], winners_total, \
(winners[0] / winners_total) * 100.0))
def random_selfplay(self):
players = [RandomAgent(Game.TOKENS[0]), RandomAgent(Game.TOKENS[1])]
game = Game.new()
game.SLEEP = 0
winner = game.play(players, draw=True)