Unsupervised machine learning with multivariate Gaussian mixture model which supports both offline data and real-time data stream.

Luka

Last update: Oct 7, 2022

Related tags

Machine Learning gaussian-mixture-model

Overview

Gaussian Mixture Model

Unsupervised machine learning with multivariate Gaussian mixture model which supports both offline data and real-time data stream.

Demo: https://lukapopijac.github.io/gaussian-mixture-model/

Installation

npm install gaussian-mixture-model

Usage

In Node.js, simply require:

const GMM = require('gaussian-mixture-model');

For browser use, include dist/gmm.js file in your project. It will create a global variable GMM.

Simple Example

// initialize model
var gmm = new GMM({
	weights: [0.5, 0.5],
	means: [[-25, 40], [-60, -30]],
	covariances: [
		[[400,0],[0,400]],
		[[400,0],[0,400]]
	]
});

// create some data points
var data = [
	[11,42],[19,45],[15,36],[25,38],[24,33],
	[-24,3],[-31,-4],[-34,-14],[-25,-5],[-16,7]
];

// add data points to the model
data.forEach(p => gmm.addPoint(p));

// run 5 iterations of EM algorithm
gmm.runEM(5);

// predict cluster probabilities for point [-5, 25]
var prob = gmm.predict([-5, 25]);  // [0.000009438559331418772, 0.000002126123537376676]

// predict and normalize cluster probabilities for point [-5, 25]
var probNorm = gmm.predictNormalize([-5, 25]);  // [0.8161537535012295, 0.18384624649877046]

License

This software is released under the MIT license.

Fork, customize and deploy your Candy Machine v2 super quickly

Candy Machine V2 Frontend This is a barebones implementation of Candy Machine V2 frontend, intended for users who want to quickly get started selling

Oct 24, 2022

Support Vector Machine (SVM) library for nodejs

node-svm Support Vector Machine (SVM) library for nodejs. Support Vector Machines Wikipedia : Support vector machines are supervised learning models t

Nov 6, 2022

Deep Learning in Javascript. Train Convolutional Neural Networks (or ordinary ones) in your browser.

ConvNetJS ConvNetJS is a Javascript implementation of Neural networks, together with nice browser-based demos. It currently supports: Common Neural Ne

Dec 31, 2022

K-nearest neighbors algorithm for supervised learning implemented in javascript

kNear Install npm install knear --save About kNear is a javascript implementation of the k-nearest neighbors algorithm. It is a supervised machine lea

Mar 7, 2022

🤖chat discord bot powered by Deep learning algorithm🧠

✨ Akaya ✨ ❗ Discord integration functionality not implemented yet! Only the deep-learning module working. Install git clone https://github.com/LyeZinh

Jun 23, 2022

JavaScript API for face detection and face recognition in the browser and nodejs with tensorflow.js

face-api.js JavaScript face recognition API for the browser and nodejs implemented on top of tensorflow.js core (tensorflow/tfjs-core) Click me for Li

Jan 2, 2023

Training Buddy has the vision of providing a secure and user-friendly way for exercise enthusiasts of all levels to be able to enhance their fitness journeys by connecting with like-minded users with similar training interests and scheduling workouts together.

Derived - Training Buddy Project Technologies Frontend Angular API Node/NestJs Backend NestJs Database PostgreSQL Hosting GCP Authentication Fir

Dec 15, 2022

architecture-free neural network library for node.js and the browser

Synaptic Important: Synaptic 2.x is in stage of discussion now! Feel free to participate Synaptic is a javascript neural network library for node.js a

Dec 27, 2022

A WebGL accelerated JavaScript library for training and deploying ML models.

TensorFlow.js TensorFlow.js is an open-source hardware-accelerated JavaScript library for training and deploying machine learning models. ⚠️ We recent

Jan 4, 2023

Comments

In the example page, implement adding more than one point at once

User should be able to add e.g. 5 points with one click. Implement this ability, and randomly create those 5 points in the area where user clicked. To create random position for a point, generate random angle and random distance from the clicked place.

opened by lukapopijac 1
Detect singluarity

When a gaussian shrinks into just one or two data points, eigenvalues of covariance matrix go to zero, and eventually we have division by zero (when calculating matrix inverse).

opened by lukapopijac 1
Don't change initial matrices given as input

Currently if you call constructor with e.g. covariance matrix as initial input value, the algorithm will have its reference and change its values in-place. Don't change given input parameters. Create their copies instead.

opened by lukapopijac 0