Aquarium: Neural Network IDE

Aquarium is a specialized IDE for designing, training, debugging, and deploying neural networks using the Neural framework. It provides a visual interface for neural network development with real-time shape propagation and error detection.

Features

• Visual Network Designer: Drag-and-drop interface for creating neural network architectures
• Real-time Shape Propagation: Visualize tensor dimensions as they flow through your network
• Integrated Debugging: Catch dimensional errors before training begins
• Neural DSL Integration: Automatically generate Neural DSL code from visual designs
• Performance Analysis: Estimate computational requirements and memory usage

Technology Stack

• Frontend: Tauri with JavaScript/HTML/CSS
• Backend: Rust for performance-critical components
• Neural Integration: Direct integration with Neural's shape propagator and other components

Development Setup

Prerequisites

• Rust
• Node.js
• Tauri CLI

Getting Started

1. Clone the repository:

   git clone https://github.com/yourusername/aquarium.git
   cd aquarium

2. Install dependencies:

   npm install

3. Run the development server:

   cargo tauri dev

Building for Production

To build the application for production:

cargo tauri build

Integration with Neural

Aquarium integrates with the Neural framework to provide:

1. Shape Propagation: Leverages Neural's shape propagator to calculate tensor dimensions
2. Code Generation: Generates Neural DSL code from visual designs
3. Training Integration: Connects with Neural's training capabilities
4. Debugging Tools: Provides insights into model performance and potential issues

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Recommended IDE Setup

• VS Code with:
  • Tauri
  • rust-analyzer