kmeans_varpar(X,k)

Implementation of K-means with Variance Partitioning initialization

Ahora está siguiendo esta publicación

Implementation of K-means with Variance Partitioning initialization. Variance Partitioning initialization is a deterministic way of initializing the data centroids, thus producing results that are repeatable and reproducible, without having to resort to tricks like seeding the pseudorandom number generator.

Citar como

Stefan Philippo Pszczolkowski Parraguez (2026). kmeans_varpar(X,k) (https://es.mathworks.com/matlabcentral/fileexchange/57229-kmeans_varpar-x-k), MATLAB Central File Exchange. Recuperado .

Agradecimientos

Inspirado por: k-means++

Información general

Compatibilidad con la versión de MATLAB

  • Compatible con cualquier versión

Compatibilidad con las plataformas

  • Windows
  • macOS
  • Linux
Versión Publicado Notas de la versión Action
1.0.1.0

Removed loop that made sure that the number of returned centrers is equal to the specified k. This is arguably not necessary and since variance partitioning provides a deterministic result, there is potential for getting trapped in an infinite loop.
Removed loop that made sure that the number of returned centrers is equal to the specified k. This is arguably not necessary and since variance partitioning provides a deterministic result, there is potential for getting trapped in an infinite loop.

1.0.0.0