pub struct ExperimentRunner { /* private fields */ }Expand description
The experiment runner. Stateless across experiments.
Implementations§
Source§impl ExperimentRunner
impl ExperimentRunner
Sourcepub fn new(config: ExperimentConfig) -> Self
pub fn new(config: ExperimentConfig) -> Self
A runner with config.
Sourcepub async fn run<F>(
&self,
dataset: &Dataset,
prepare: F,
) -> Result<ExperimentReport>
pub async fn run<F>( &self, dataset: &Dataset, prepare: F, ) -> Result<ExperimentReport>
Run prepare’s agent over every case of dataset,
runs_per_case times each, and grade the evidence.
prepare builds the graph under test for one case run and receives
the run’s journal for evidence wiring. At most the configured number
of runs are in flight. The default sequential path retains fail-fast
behavior. In parallel mode, observing an infrastructure error stops
admission of new runs; the already-active window settles, and the
earliest dataset-case/repetition error from that window wins
deterministically.
Auto Trait Implementations§
impl !RefUnwindSafe for ExperimentRunner
impl !UnwindSafe for ExperimentRunner
impl Freeze for ExperimentRunner
impl Send for ExperimentRunner
impl Sync for ExperimentRunner
impl Unpin for ExperimentRunner
impl UnsafeUnpin for ExperimentRunner
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