def __init__(self, num_holes, holes_to_consider, time_to_consider, name):
pick_agent.init(self, num_holes, name)
self.holes_to_consider = holes_to_consider
self.time_to_consider = time_to_consider
self.steps = zip(time_to_consider, holes_to_consider)
self.rounds = 0
self.step_count = 0
self.current_step = 0
def __init__(self, num_holes, name, init_Q = 0.5, eta=0.1, step_size=None):
self.Q = np.ones(num_holes) * init_Q
self.action_counts = np.zeros(num_holes)
self.action_sums = np.zeros(num_holes)
self.eta = eta
self.step_size = None
pick_agent.init(self, num_holes, name)
def __init__(self, num_holes, name, init_Q = 0.5, temp=5.0, step_size=None):
self.Q = np.ones(num_holes) * init_Q
self.action_counts = np.zeros(num_holes)
self.action_sums = np.zeros(num_holes)
self.step_size = step_size
self.temp = temp
pick_agent.init(self, num_holes, name)
def __init__(self, num_holes, name):
pick_agent.init(self, num_holes, name)
