Machine Learning Algorithms Visualizations - MLens

MLens is an interactive web application for visualizing machine learning algorithms, built with Three.js and React. This application serves as a supplementary learning material for students interested in machine learning, providing intuitive visual understanding of selected algorithms.

Live Demo

Visit https://mlens.wittlinger.dev to see the application in action.

Features

• Interactive 2D and 3D visualizations of machine learning algorithms
• Step-by-step execution with detailed explanations
• Configurable algorithm parameters
• Mobile-friendly responsive design
• Currently implemented algorithms:
  • K-means Clustering
  • Linear Regression
  • Support Vector Machines (SVM)
  • Feed Forward Neural Network (FFNN)
  • Autoencoder
  • XGBoost

Technology Stack

• TypeScript
• React
• Next.js
• Three.js
• React Three Fiber
• TailwindCSS
• ShadCN UI
• Docker

Getting Started

Prerequisites

• Docker (v20.10.0 or higher)

Installation and Running

Make sure that ports 3000 and 8000 are available on your machine.

1. Navigate to the root of the repository

2. Build and run with Docker

   docker compose up

3. Open http://localhost:3000 in your browser

Project Structure

apps/
└── web/                         # Next.js web application
└── api/                         # Supplementary Python API
packages/
    ├── algorithms/              # Algorithm implementations and core logic
    ├── eslint-config/           # ESLint configuration files
    ├── typescript-config/       # TypeScript configuration files
    └── ui/                      # Reusable UI components and hooks

Adding New Algorithms

The application is designed to be extensible. To add a new algorithm with a certain {slug}

1. Create a new file for the algorithm in packages/algorithms/impl/{slug}/{slug}
2. Implement the AlgorithmDefinition interface
3. Add algorithm's metadata to the registry in packages/algorithms/impl/index.ts

See existing algorithm implementations for examples.

License

This project is licensed under the BSD 3-Clause License - see the LICENSE file for details.

Related

This application was created as part of a bachelor's thesis at Faculty of Informatics, Masaryk University.