[source]

K Means

A fast online centroid-based hard clustering algorithm capable of grouping linearly separable data points given some prior knowledge of the target number of clusters (defined by k). K Means is trained using adaptive Mini Batch Gradient Descent and minimizes the inertia cost function at each epoch. Inertia is defined as the average sum of distances between each sample and its nearest cluster centroid.

Interfaces: Estimator, Learner, Online, Probabilistic, Persistable, Verbose

Data Type Compatibility: Continuous

Parameters

#	Name	Default	Type	Description
1	k		int	The number of target clusters.
2	batchSize	128	int	The size of each mini batch in samples.
3	epochs	300	int	The maximum number of training rounds to execute.
4	minChange	1e-4	float	The minimum change in the inertia for training to continue.
5	kernel	Euclidean	Distance	The distance kernel used to compute the distance between sample points.
6	seeder	PlusPlus	Seeder	The seeder used to initialize the cluster centroids.

Example

use Rubix\ML\Clusterers\KMeans;
use Rubix\ML\Kernels\Distance\Euclidean;
use Rubix\ML\Clusterers\Seeders\PlusPlus;

$estimator = new KMeans(3, 128, 300, 10.0, new Euclidean(), new PlusPlus());

Additional Methods

Return the k computed centroids of the training set.

public centroids() : array[]

Return the number of training samples that each centroid is responsible for.

public sizes() : int[]

Return the loss for each epoch from the last training session.

public losses() : float[]|null

Return the progress table combining every epoch recorded during the last training session — the inertia at each epoch — into a single ordered sequence.

public progress() : iterable

References

---

1. D. Sculley. (2010). Web-Scale K-Means Clustering. ↩