core.py 文件源码

def run_episode(self, env: gym.Env):

        meta_wrapper = get_wrapper(env, chi.rl.wrappers.Wrapper)

        done = False
        ob = env.reset()
        a, meta = self.act(ob)

        rs = []
        while not done:
            if meta_wrapper:
                meta_wrapper.set_meta(meta)  # send meta information to wrappers
            ob, r, done, info = env.step(a)
            a, meta = self.act(ob, r, done, info)
            rs.append(r)

        return sum(rs)