kmeans-same-size
v1.0.1
Published
A implementation of a variant of k-means algorithm to get groups from the waypoints with the same size
Downloads
110
Readme
kmeans-same-size
This a variation of k-means algorithm with the same size for each group oriented to waypoints with a distribution heterogeneously on a city.
Installation
$ npm install kmeans-same-size
Example
const KmeansLib = require('kmeans-same-size');
var kmeans = new KmeansLib();
const k = 10; // Groups Number
const size = 3 // Group size
let vectors = [
{ x: -1.10, y: -1.10 },
{ x: -1.20, y: -1.20 },
{ x: -1.30, y: -1.30 },
{ x: -1.40, y: -1.40 },
{ x: -1.50, y: -1.50 },
{ x: -1.60, y: -1.60 },
{ x: -1.70, y: -1.70 },
{ x: -1.80, y: -1.80 },
{ x: -1.90, y: -1.90 },
{ x: -1.91, y: -1.92 },
{ x: -1.93, y: -1.93 }
]
kmeans.init({k: k, runs: size, equalSize: true, normalize: false });
const sum = kmeans.calc(vectors);
The vector is mutated
console.log(vectors);