Struct MiniBatchKMeans 

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

Mini-Batch K-Means clustering.

Approximates standard K-Means by updating centroids using random mini-batches of the data at each iteration, rather than the full dataset. This is significantly faster for large datasets while producing similar results.

Implementations

impl MiniBatchKMeans

pub fn new(k: usize) -> Self

Create a Mini-Batch K-Means model with k clusters.

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

Set the mini-batch size (default 1024).

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

Set maximum iterations.

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

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

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

Predict cluster assignments for new data.

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 run.

Trait Implementations

impl Clone for MiniBatchKMeans

fn clone(&self) -> MiniBatchKMeans

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl PartialFit for MiniBatchKMeans

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

Update centroids with a streaming average over the given batch.

On the first call, initializes centroids via K-Means++ on the batch. Subsequent calls assign each sample to the nearest centroid and update it with a decaying learning rate (1 / count).

fn is_initialized(&self) -> bool

Whether the model has been initialized (at least one partial_fit call).