Struct border_core::Step
source · pub struct Step<E: Env> {
pub act: E::Act,
pub obs: E::Obs,
pub reward: Vec<f32>,
pub is_done: Vec<i8>,
pub info: E::Info,
pub init_obs: E::Obs,
}Expand description
Represents an action, observation and reward tuple (a_t, o_t+1, r_t)
with some additional information.
An environment emits Step object at every interaction steps.
This object might be used to create transitions (o_t, a_t, o_t+1, r_t).
Old versions of the library support veectorized environments, which requires
elements in Step to be able to handle multiple values.
This is why reward and is_done are vector.
Fields§
§act: E::ActAction.
obs: E::ObsObservation.
reward: Vec<f32>Reward.
is_done: Vec<i8>Flag denoting if episode is done.
info: E::InfoInformation defined by user.
init_obs: E::ObsInitial observation. If is_done[i] == 0, the corresponding element will not be used.
Implementations§
Auto Trait Implementations§
impl<E> RefUnwindSafe for Step<E>where <E as Env>::Act: RefUnwindSafe, <E as Env>::Info: RefUnwindSafe, <E as Env>::Obs: RefUnwindSafe,
impl<E> Send for Step<E>where <E as Env>::Act: Send, <E as Env>::Info: Send, <E as Env>::Obs: Send,
impl<E> Sync for Step<E>where <E as Env>::Act: Sync, <E as Env>::Info: Sync, <E as Env>::Obs: Sync,
impl<E> Unpin for Step<E>where <E as Env>::Act: Unpin, <E as Env>::Info: Unpin, <E as Env>::Obs: Unpin,
impl<E> UnwindSafe for Step<E>where <E as Env>::Act: UnwindSafe, <E as Env>::Info: UnwindSafe, <E as Env>::Obs: UnwindSafe,
Blanket Implementations§
source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere T: ?Sized,
source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more