deck = np.arange(-5, 11) done = False api = Api() actionspace = np.arange(1, 16) while not done: random_action = 0 while random_action == 0: random_action = np.random.choice(actionspace) actionspace[random_action - 1] = 0 env_action = 0 while env_action == 0: env_action = np.random.choice(deck) deck[env_action + 5] = 0 done, bot_action, bot_score, random_score = api.do_action( random_action, env_action) print("Bot Score: " + str(bot_score)) print("Random policy Score: " + str(random_score)) print("") if all(a == 0 for a in actionspace): done = True if bot_score > random_score: botscore += 1 if bot_score < random_score: randomscore += 1 if bot_score == random_score: draw += 1 print("") print("Bot: " + str(botscore))