def __init__(self, env_name, num_episodes, alpha, gamma, policy, report_freq=100, **kwargs):
"""
base class for RL using lookup table
:param env_name: see https://github.com/openai/gym/wiki/Table-of-environments
:param num_episodes: int, number of episode for training
:param alpha: float, learning rate
:param gamma: float, discount rate
:param policy: str
:param report_freq: int, by default 100
:param kwargs: other arguments
"""
self.env = gym.make(env_name)
self.num_episodes = num_episodes
self.alpha = alpha
self.gamma = gamma
self.state = None
self._rewards = None
self._policy = policy
self.report_freq = report_freq
for k, v in kwargs.items():
setattr(self, str(k), v)
评论列表
文章目录