An easy visual way to design neural networks

BlockDL is a web-based visual editor for building deep learning models. Drag and drop layers, connect them visually, and generate Keras code automatically.

This repository contains the code for the core code-generation interface and engine. The courses on htttps://blockdl.com are yet to be open-sourced.

• 🎨 Visual Design: Drag and drop neural network layers
• 🔗 Smart Connections: Automatic shape validation, so you catch problems early
• 🐍 Code Generation: Copy working Keras code instantly as you build your network
• 🚀 No Installation: Runs entirely in your browser
• ⚡ High Performance: Optimized with code splitting, memoization, and asset optimization
• 📊 Performance Monitoring: Real-time Web Vitals tracking and bundle analysis
• 🎯 Production Ready: Comprehensive testing, error handling, and monitoring

Visit https://blockdl.com to start designing neural networks immediately.

1. Clone the repository

   git clone https://github.com/Aryagm/blockdl.git
   cd blockdl

2. Install dependencies

   npm install

3. Start the development server

   npm run dev

4. Open your browser Navigate to http://localhost:5173

• Bundle Analysis: Run npm run analyze to visualize bundle composition
• Code Splitting: Automatic route and component-level splitting for faster loading
• Asset Optimization: Compressed images, CSS, and optimized fonts
• React Memoization: Comprehensive component optimization to prevent unnecessary re-renders
• Performance Monitoring: Real-time Web Vitals tracking with development dashboard

npm run dev              # Start development server
npm run build            # Production build
npm run analyze          # Bundle size analysis
npm run analyze:ci       # CI-friendly analysis with monitoring
npm run optimize:analyze # Full optimization analysis

• BUNDLE_ANALYSIS_GUIDE.md - Bundle optimization guide
• ASSET_OPTIMIZATION.md - Asset optimization documentation
• PERFORMANCE_MONITORING.md - Performance monitoring setup
• PRODUCTION_READINESS_PLAN.md - Production deployment guide

This project is licensed under the Mozilla Public License 2.0 - see the LICENSE.md file for details.