https://github.com/infinityfx-llc/ml-basic

Lightweight, zero dependency, machine learning library.
https://github.com/infinityfx-llc/ml-basic

ai classifier convolutional javascript lightweight machine-learning ml multithreading network neural npm recurrent zero-dependency

Last synced: 5 months ago
JSON representation

Lightweight, zero dependency, machine learning library.

• Host: GitHub
• URL: https://github.com/infinityfx-llc/ml-basic
• Owner: infinityfx-llc
• License: gpl-3.0
• Created: 2022-05-11T11:43:28.000Z (about 4 years ago)
• Default Branch: main
• Last Pushed: 2025-07-29T19:32:51.000Z (11 months ago)
• Last Synced: 2025-09-01T19:32:00.541Z (10 months ago)
• Topics: ai, classifier, convolutional, javascript, lightweight, machine-learning, ml, multithreading, network, neural, npm, recurrent, zero-dependency
• Language: TypeScript
• Homepage:
• Size: 241 KB
• Stars: 2
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# ml-basic

[![NPM package](https://img.shields.io/npm/v/ml-basic)](https://www.npmjs.com/package/ml-basic)

![NPM bundle size](https://img.shields.io/bundlephobia/minzip/ml-basic)

![Last commit](https://img.shields.io/github/last-commit/infinityfx-llc/ml-basic)

![NPM weekly downloads](https://img.shields.io/npm/dw/ml-basic)

![NPM downloads](https://img.shields.io/npm/dt/ml-basic)

Lightweight, zero dependency, machine learning library for use in NodeJS and browsers.

## Table of contents

- [Importing](#importing)

- [Usage](#usage)

- [Training](#training)

## Importing

### CommonJS

```javascript

const MLBasic = require('ml-basic');

```

### ESM

```javascript

import MLBasic from 'ml-basic';

```

### Browser

```html

```

## Usage

```javascript

const net = new Neural({

    layers: [

        ...

    ]

});

const result = net.predict([1, 0]);

// result = [0.532..]

```

## Training

```javascript

import { DataFrame } from 'ml-basic';

const data = new DataFrame([ ... ]);

const error = await net.fit(data);

// error = 0.532..

```