Struct Hdbscan 

#[non_exhaustive]
pub struct Hdbscan { /* private fields */ }

HDBSCAN clustering model.

Automatically determines the number of clusters from density variations in the data. Noise points are labeled -1.

Implementations

impl Hdbscan

pub fn new() -> Self

Create a new HDBSCAN model with default parameters.

Default: min_cluster_size = 5, min_samples = min_cluster_size.

pub fn min_cluster_size(self, size: usize) -> Self

Set the minimum cluster size (default: 5).

Clusters with fewer points than this are dissolved into noise.

pub fn min_samples(self, k: usize) -> Self

Set min_samples for core distance computation.

Default: same as min_cluster_size. Smaller values produce more clusters; larger values make the algorithm more conservative.

pub fn metric(self, m: DistanceMetric) -> Self

Set the distance metric (default: DistanceMetric::Euclidean).

Use DistanceMetric::Cosine for L2-normalized embeddings (e.g. face embeddings, CLIP vectors).

pub fn labels(&self) -> &[i32]

Get cluster labels (-1 = noise).

pub fn n_clusters(&self) -> usize

Number of clusters found (excluding noise).

pub fn n_noise(&self) -> usize

Number of noise points.

pub fn outlier_scores(&self) -> &[f64]

Per-point outlier scores (higher = more outlier-like).

Based on the ratio of the point’s lambda value to its cluster’s lambda birth. Points deep inside clusters have scores near 0; borderline points have scores near 1.

pub fn fit(&mut self, data: &Dataset) -> Result<()>

Fit the HDBSCAN model on a dataset.

pub fn fit_predict(&mut self, data: &Dataset) -> Result<Vec<i32>>

Convenience: fit and return labels.

Trait Implementations

impl Clone for Hdbscan

fn clone(&self) -> Hdbscan

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Hdbscan

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

Formats the value using the given formatter.

impl Default for Hdbscan

fn default() -> Self

Returns the “default value” for a type.