Playing the game of Prisoners Dilemma

Prisoners2.py

class Game:
    def __init__(self, max_game=100):
        self.p1 = Player('Agent A')
        self.p2 = Player('Agent B')
        self.max_game = max_game
        
    def play(self, avg_regret_matching=False):
        def play_regret_matching():
            for i in xrange(0, self.max_game):
                self.p1.update_strategy()
                self.p2.update_strategy()
                a1 = self.p1.action()
                a2 = self.p2.action()
                self.p1.regret(a1, a2)
                self.p2.regret(a2, a1)

                acts = (a1, a2)
                num_wins[acts] += 1

        def play_avg_regret_matching():
            for i in xrange(0, self.max_game):
                a1 = self.p1.action(use_avg=True)
                a2 = self.p2.action(use_avg=True)
                acts = (a1, a2)
                num_wins[acts] += 1

        num_wins = {
                acts:0
                for acts in itertools.product(
                        Prisoners.actions,
                        Prisoners.actions
                )
            }

        play_regret_matching() if not avg_regret_matching else play_avg_regret_matching()
        print num_wins

    def conclude(self):
        """
        let two players conclude the average strategy from the previous strategy stats 
        """
        self.p1.learn_avg_strategy()
        self.p2.learn_avg_strategy()