Struct KMeans 

pub struct KMeans {
    pub centroids: Vec<Vec<f32>>,
    pub dimension: usize,
    pub iterations: usize,
    pub inertia: f32,
}

K-means clustering result.

Fields

centroids: Vec<Vec<f32>>

The cluster centroids.

dimension: usize

The dimension of each centroid.

iterations: usize

Number of iterations run.

inertia: f32

Final inertia (sum of squared distances to nearest centroid).

Implementations

impl KMeans

pub fn train( data: &[&[f32]], config: &KMeansConfig, metric: DistanceMetric, ) -> Result<Self, VectorError>

Train k-means on the given data.

Arguments

• data: Training vectors, each with the same dimension
• config: K-means configuration
• metric: Distance metric for clustering

Errors

Returns an error if:

• data is empty
• Vectors have inconsistent dimensions
• k is greater than the number of data points

pub fn find_nearest(&self, vector: &[f32], metric: DistanceMetric) -> usize

Find the index of the nearest centroid to the given vector.

Trait Implementations

impl Clone for KMeans

fn clone(&self) -> KMeans

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for KMeans

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

Formats the value using the given formatter.