Enum linfa_clustering::KMeansInit

#[non_exhaustive]
pub enum KMeansInit<F: Float> {
    Random,
    Precomputed(Array2<F>),
    KMeansPlusPlus,
    KMeansPara,
}

Specifies centroid initialization algorithm for KMeans.

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

Random

Pick random points as centroids.

Precomputed(Array2<F>)

Precomputed list of centroids, represented as an array of (n_centroids, n_features).

KMeansPlusPlus

K-means++ algorithm. Using this over random initialization causes K-means to converge faster for almost all cases, since K-means++ produces better centroids.

KMeansPara

K-means|| algorithm, a parallelized version of K-means++. Performs much better than K-means++ when the number of clusters is large (>100) while producing similar centroids, so use this for larger datasets. Details on the algorithm can be found here.

Trait Implementations

impl<F: Clone + Float> Clone for KMeansInit<F>

fn clone(&self) -> KMeansInit<F>

Returns a copy of the value.

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

Performs copy-assignment from source.

impl<F: Debug + Float> Debug for KMeansInit<F>

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

Formats the value using the given formatter.

impl<F: PartialEq + Float> PartialEq for KMeansInit<F>

fn eq(&self, other: &KMeansInit<F>) -> bool

This method tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

This method tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl<F: Float> StructuralPartialEq for KMeansInit<F>