mtralka / GeoBuckets

Lightweight TypeScript library for data classification

🌎🪣 GeoBuckets

Lightweight Javascript (TypeScript, really) library for classification. Dependency free, tested, and type safe.

Features

• 🔥 supports 6 classification types
• ⚖️: Small!
  • 🚀 4.4kb minifed, 1.7kb gzipped
  • 0️⃣ zero dependencies
  • 🌲 tree-shakeable
• 🧪 fully tested
• 🏷️ fully typed
• 🪨 asynchronous and non-blocking

Supports

• Equal Intervals (EQI)
  • equalIntervalBuckets
• Standard Deviation (STD)
  • standardDeviationBuckets
• Arithmetic Progression (APG)
  • arithmeticProgressionBuckets
• Geometric Progression (GPG)
  • geometricProgressionBuckets
• Quantiles (QNT)
  • quantileBuckets
• Jenks (JNK)
  • jenksBuckets

Install

Install through NPM

npm install geobuckets

Usage

All outward facing functions are async and must be called with await XXX

Explicit import:

import { jenksBuckets } from "geobuckets";

const data: Array<number> = [60, 26, 20, 17, 10, 27, 98, 42, 55, 35];
const numClasses: number = 3;

const buckets: Array<number> = await jenksBuckets(data, numClasses);

console.log(buckets) >> [10, 35, 60, 98];

Helper function and type

import { generateBuckets, BucketTypes } from "geobuckets";

const data: Array<number> = [60, 26, 20, 17, 10, 27, 98, 42, 55, 35];
const numClasses: number = 3;

const buckets: Array<number> = await generateBuckets(
  BucketTypes.JNK,
  data,
  numClasses
);

console.log(buckets) >> [10, 35, 60, 98];

A modern implementation of the GeoStats library by Simon Georget (MIT license)