Struct KMeans 

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

K-Means clustering.

Uses k-means++ initialization for better convergence. When n_init > 1 (default 10), the algorithm runs multiple times with different random seeds and keeps the result with the lowest inertia.

Implementations

impl KMeans

pub fn new(k: usize) -> Self

Create a K-Means model with k clusters.

pub fn max_iter(self, n: usize) -> Self

Set maximum iterations per run.

pub fn tolerance(self, t: f64) -> Self

Set convergence tolerance.

pub fn tol(self, t: f64) -> Self

Alias for tolerance (sklearn convention).

pub fn seed(self, s: u64) -> Self

Set random seed.

pub fn n_init(self, n: usize) -> Self

Set the number of independent runs with different random seeds.

The result with the lowest inertia is kept. Default is 10, matching sklearn. Set to 1 for a single run (faster but less reliable).

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

Fit the model on a dataset (uses features only, ignores target).

When n_init > 1, runs K-Means multiple times and keeps the best.

pub fn predict(&self, features: &[Vec<f64>]) -> Result<Vec<usize>>

Predict cluster assignments for new data.

pub fn transform(&self, features: &[Vec<f64>]) -> Result<Vec<Vec<f64>>>

Transform data into cluster-distance space.

Returns a n_samples × k matrix where each value is the Euclidean distance from the sample to each centroid.

Example

let distances = km.transform(&[vec![5.0, 5.0]]).unwrap();
assert_eq!(distances[0].len(), 2); // one distance per centroid

pub fn centroids(&self) -> &[Vec<f64>]

Get the cluster centroids.

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

Get cluster labels for training data.

pub fn inertia(&self) -> f64

Sum of squared distances to the nearest centroid.

pub fn n_iter(&self) -> usize

Number of iterations to converge.

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 Tunable for KMeans

fn set_param(&mut self, name: &str, value: ParamValue) -> Result<()>

Apply a named hyperparameter.

fn clone_box(&self) -> Box<dyn Tunable>

Clone this model into a boxed trait object.

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

Train on a dataset.

fn predict(&self, features: &[Vec<f64>]) -> Result<Vec<f64>>

Predict on row-major features.