BufferedTransition
One n-step transition emitted by an NStepBuffer, staged for the replay buffer.
Values
action
python
action: intdone
python
done: booln_steps
python
n_steps: intnext_state
python
next_state: np.ndarrayreward
python
reward: floatreward_shaping_metadata
python
reward_shaping_metadata: dict[str, int | float] | Nonestate
python
state: np.ndarray