Struct Optics

pub struct Optics<A, M> {
    pub eps: A,
    pub min_samples: usize,
    pub metric: M,
    /* private fields */
}

OPTICS (ordering points to identify the clustering structure) clustering algorithm.

Examples

use ndarray::array;
use petal_neighbors::distance::Euclidean;
use petal_clustering::{Optics, Fit};

let points = array![[1., 2.], [2., 5.], [3., 6.], [8., 7.], [8., 8.], [7., 3.]];
let clustering = Optics::new(4.5, 2, Euclidean::default()).fit(&points, None);

assert_eq!(clustering.0.len(), 2);        // two clusters found
assert_eq!(clustering.0[&0], [0, 1, 2]);  // the first three points in Cluster 0
assert_eq!(clustering.0[&1], [3, 4, 5]);  // the rest in Cluster 1

Fields

eps: A

The radius of a neighborhood.

min_samples: usize

The minimum number of points required to form a dense region.

metric: M

The metric to compute distance between the entries.

Implementations

impl<A, M> Optics<A, M>

where A: FloatCore, M: Metric<A>,

pub fn new(eps: A, min_samples: usize, metric: M) -> Self

pub fn extract_clusters_and_outliers( &self, eps: A, ) -> (HashMap<usize, Vec<usize>>, Vec<usize>)

Trait Implementations

impl<A: Debug, M: Debug> Debug for Optics<A, M>

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

Formats the value using the given formatter.

impl<A> Default for Optics<A, Euclidean>

where A: FloatCore,

fn default() -> Self

Returns the “default value” for a type.

impl<'de, A, M> Deserialize<'de> for Optics<A, M>

where A: Deserialize<'de>, M: Deserialize<'de>,

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl<S, A, M> Fit<ArrayBase<S, Dim<[usize; 2]>>, HashMap<usize, Vec<usize>>, (HashMap<usize, Vec<usize>>, Vec<usize>)> for Optics<A, M>

where A: AddAssign + DivAssign + FloatCore + FromPrimitive + Send + Sync, S: Data<Elem = A> + Sync, M: Metric<A> + Clone + Sync,

Fits the OPTICS clustering algorithm to the given input data.

Parameters

• input: A 2D array representing the dataset to cluster. Each row corresponds to a data point.
• _params: An optional parameter for prelabelled data. Not used in this implementation, but required for consistency.

Returns

A tuple containing:

• HashMap<usize, Vec<usize>>: A mapping of cluster IDs to the indices of points in each cluster.
• Vec<usize>: A vector of indices representing the noise points that do not belong to any cluster.

fn fit( &mut self, input: &ArrayBase<S, Ix2>, _params: Option<&HashMap<usize, Vec<usize>>>, ) -> (HashMap<usize, Vec<usize>>, Vec<usize>)

impl<A, M> Serialize for Optics<A, M>

where A: Serialize, M: Serialize,

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.