def reset(self, **kwargs):
if (self._init_reset and self.error_on_premature_reset
and self.n_transitions > 0): # noqa: E127
raise RuntimeError(
"BufferingWrapper reset() before samples were accessed")
self._init_reset = True
self.n_transitions = 0
obs = self.venv.reset(**kwargs)
self._traj_accum = rollout.TrajectoryAccumulator()
for i, ob in enumerate(obs):
self._traj_accum.add_step({"obs": ob}, key=i)
return obs
def reset(self) -> np.ndarray:
"""Resets the environment.
Returns:
obs: first observation of a new trajectory.
"""
self.traj_accum = rollout.TrajectoryAccumulator()
obs = self.env.reset()
self._last_obs = obs
self.traj_accum.add_step({"obs": obs})
self._done_before = False
self._is_reset = True
return obs