def print_policy_table()

in simple_game/load_strategies.py [0:0]

• 30 lines of code
• 10 McCabe index (conditional complexity)

def print_policy_table(policy):
    # Get max c1 and c2
    maxcards = [0, 0]
    for (player, card, history), p in policy.items():
        maxcards[player - 1] = max(maxcards[player - 1], card) 

    table = []
        
    for c1 in range(maxcards[0] + 1):
        this_table = dict()
        for c2 in range(maxcards[1] + 1):
            history = "r"
            player = 1
            while True:
                c = c1 if player == 1 else c2
                key = (player, c, history)

                p = policy.get(key, None)

                if p is None:
                    break

                opt_a = p[1]
                history += str(opt_a)
                player = 3 - player

            if history[-1] == 'r':
                reward = 0
            else:
                contract = 1 << (int(history[-2]) - 1)
            if c1 + c2 >= contract:
                reward = contract
            else:
                reward = 0

            this_table[c2] = history[1:] + " (" + str(reward) + ")"
        table.append(this_table)

    print(pd.DataFrame(table))