invisal / noobjs

Javascript Algorithms Toolsets

NoobJS

NoobJS contains Javascript implementations of many algorithms.

Implemented Algorithms

Machine Learning

• Linear Regression
• KNN Classifier
• KMean Clustering

Graph

• Graph Data Structure
• Prim's algorithm
• Kruskal's algorithm
• Breadth-first search
• Depth-first search
• Graph Coloring using Greedy

Math

• BigNumber
• Matrix Multiplication (Brute Force)
• Matrix Inverse (Gaussian Elimination)

Elementary Data Structure

• Disjoint Set
• Queue
• Stack
• Priority Queue (Binary Heap)
• Hash Priority Queue

Tree

• Ball Tree

Random Number

• Linear Congruential Generator
• Marsaglia Polar Method

Other

• QuickSelect
• Partition (Lomuto & Hoare)
• Fisher Yates Shuffle
• Sieve of Eratosthenes

Examples

Using Linear Regression

var { LinearRegression } = noobjs;

var x = [ 
    [0.17], [2.47], [1.73], [4.43], [4.51], 
    [4.59], [1.24], [0.84], [0.09], [0.51], 
    [2.06], [2.89], [0.81], [4.39], [1.20] 
];

var y = [ 
    [100.0], [410.0], [250.0], [560.0], [440.0], 
    [510.0], [230.0], [170.0], [100.0], [140.0], 
    [250.0], [460.0], [160.0], [490.0], [210.0] 
];

var model = new LinearRegression();
model.fit(x, y);
console.log( model.predict( [ [4.0], [2.0], [1.5] ] ) );

Using Elementary Data Structures

var { PriorityQueue, Stack, Queue } = noobjs;

var p = new PriorityQueue();
p.push(7);
p.push(8);
p.push(6);

console.log(p.pop());  // 6
console.log(p.pop());  // 7
console.log(p.pop());  // 8

Using NoobJS to solve linear system of equations

var { LinearAlgebra } = noobjs;

//  x + 2y -  z = 2
// 2x + 2y + 2z = 12
//  x -  y + 2z = 5
var a = [
  [ 1,  2,  -1],
  [ 2,  2,   2],
  [ 1, -1,   2]
];

var b = [ [2], [12], [5] ];

console.log(LinearAlgebra.mat_multiply(LinearAlgebra.mat_inverse(a), b));
// [ 1.00, 2.00, 3.00 ]