pub struct Bocpd {
pub expected_run_length: f64,
pub map_drop_min: usize,
pub mu0: f64,
pub kappa0: f64,
pub alpha0: f64,
pub beta0: f64,
}Fields§
§expected_run_length: f64Expected run length. hazard = 1 / expected_run_length. 100 is
a common default in the BOCPD literature for traces of a few
hundred samples; we keep it here so the crate’s bake-off lines up
with published benchmarks.
map_drop_min: usizeMinimum drop in MAP run length that triggers a change-point. 1 would fire on every slight argmax jitter; 2 damps that without missing real regime changes.
mu0: f64Normal-Gamma prior hyper-parameters. Weakly informative defaults match Adams & MacKay’s Figure 3 setup.
kappa0: f64§alpha0: f64§beta0: f64Trait Implementations§
Source§impl ChangePointDetector for Bocpd
impl ChangePointDetector for Bocpd
Auto Trait Implementations§
impl Freeze for Bocpd
impl RefUnwindSafe for Bocpd
impl Send for Bocpd
impl Sync for Bocpd
impl Unpin for Bocpd
impl UnsafeUnpin for Bocpd
impl UnwindSafe for Bocpd
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