| 37 | |
| 38 | |
| 39 | class TimeLimit(gym.Wrapper): |
| 40 | def __init__(self, env, max_episode_steps=None): |
| 41 | super(TimeLimit, self).__init__(env) |
| 42 | self._max_episode_steps = max_episode_steps |
| 43 | self._elapsed_steps = 0 |
| 44 | |
| 45 | def step(self, ac): |
| 46 | observation, reward, done, info = self.env.step(ac) |
| 47 | self._elapsed_steps += 1 |
| 48 | if self._elapsed_steps >= self._max_episode_steps: |
| 49 | done = True |
| 50 | info['TimeLimit.truncated'] = True |
| 51 | return observation, reward, done, info |
| 52 | |
| 53 | def get_max_episode_steps(self): |
| 54 | return self._max_episode_steps |
| 55 | |
| 56 | def reset(self, **kwargs): |
| 57 | self._elapsed_steps = 0 |
| 58 | return self.env.reset(**kwargs) |
| 59 | |
| 60 | |
| 61 | class NoopResetEnv(gym.Wrapper): |