Struct BinarySegmentation 

pub struct BinarySegmentation {
    pub max_changepoints: usize,
    pub threshold: f64,
    pub min_size: usize,
    pub cost_fn: CostFunction,
}

Binary Segmentation for change point detection.

Greedy algorithm that recursively splits segments at points of maximum change statistic. Time complexity: O(n log n) average, O(n^2) worst case.

Algorithm

1. Compute test statistic for all candidate split points across the full series.
2. If the maximum statistic exceeds the threshold, accept the split point.
3. Recursively apply to each resulting sub-segment.
4. Stop when no segment can be split or max_changepoints is reached.

Fields

max_changepoints: usize

Maximum number of change points to detect

threshold: f64

Threshold for accepting a change point

min_size: usize

Minimum segment size

cost_fn: CostFunction

Cost function

Implementations

impl BinarySegmentation

pub fn new(max_changepoints: usize, threshold: f64, min_size: usize) -> Self

Create a new binary segmentation detector.

pub fn with_cost( max_changepoints: usize, threshold: f64, min_size: usize, cost_fn: CostFunction, ) -> Self

Create with a specific cost function.

pub fn detect(&self, data: &ArrayView1<'_, f64>) -> Result<ChangePointResult>

Detect change points in the data.

Trait Implementations

impl Clone for BinarySegmentation

fn clone(&self) -> BinarySegmentation

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for BinarySegmentation

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.