package prbnmcn-clustering

You can search for identifiers within the package.

in-package search v0.2.0

On This Page

Parameters
Signature

prbnmcn-clustering
- CHANGES
- LICENSE
- README
- Library prbnmcn-clustering
  - Clustering
    
    Agglomerative
    
    Element_set
    
    Make
    
    Helpers
    
    Intf
    
    Metric
    
    K_means
    
    Element
    
    Make
    
    K_medoids
    
    Make
    
    MultiStart
- Sources
  - prbnmcn-clustering
    
    agglomerative.ml
    
    clustering.ml
    
    helpers.ml
    
    intf.ml
    
    k_means.ml
    
    k_medoids.ml
    
    multiStart.ml

Legend:
Page
Library
Module
Module type
Parameter
Class
Class type
Source

Module `K_means.Make`Source

Parameters

module E : Element

Signature

Source

val k_means : 
  k:int ->
  init:init ->
  elements:E.t array ->
  termination:termination ->
  Random.State.t ->
  E.t array array

k_means performs the clustering using to the provided initialization method.

Sourceval cost : classes:E.t array array -> float

cost returns the sum over all classes of the sum of squared distances from the mean of the class to all elements of the class. This quantity will decrease monotonically as k increases, and the usual method to determine k is to select the one where this quantity's decrease has its first inflection.

package prbnmcn-clustering

Module K_means.MakeSource

Parameters

Signature

Module `K_means.Make`Source