def __init__ (self, size_x, size_y, beta, hidden, learning_rate, reward): Bot.__init__(self) self.bot_name = "Bot_RL_MLP" self.mlp = MLP (size_x * size_y, hidden, size_x * size_y, learning_rate) self.reward = reward[:] #hoher Wert für beta (50?): exploitation #niedriger Wert für beta : exploration self.beta = beta
def __init__ (self, size_x = 3, size_y = 3, beta = 1, hidden = 20, learning_rate = 0.1, reward = [0, 1.0, -1.0], initial_field = [0], player_ID = 1): Bot.__init__(self) self.initial_field = initial_field self.player_ID = player_ID self.bot_name = "Bot_RL_MLP" self.version = 1 self.counter = 0 self.optimization = [] self.reward = reward[:] self.first_action = True self.beta = beta #hoher Wert für beta (50?): exploitation #niedriger Wert für beta : exploration self.mlp = MLP (size_x * size_y, hidden, size_x * size_y, learning_rate) self.new_game()