Deploy Safe: Local RAG System with Chinese Semantic Search

Table of Contents

• Overview
• Features
• Technologies Used
• Installation
• Usage
• Configuration
• Deployment
• Contributing
• License
• Contact

Overview

Deploy Safe is a local Retrieval-Augmented Generation (RAG) system designed for efficient Chinese semantic search. This project integrates an Ollama-based Large Language Model (LLM) with Nginx hosting and DuckDNS access. It provides a seamless experience for users looking to leverage AI capabilities in a local environment.

You can find the latest releases and download the necessary files here.

Features

• Local RAG System: Utilize a powerful retrieval system to enhance your AI applications.
• Chinese Semantic Search: Efficiently search through Chinese text with high accuracy.
• Ollama-based LLM: Leverage state-of-the-art language models for your applications.
• Nginx Hosting: Host your application with a reliable web server.
• DuckDNS Access: Easily access your application from anywhere using DuckDNS.

Technologies Used

This project incorporates a variety of technologies to deliver its functionality:

• Chatbot: A conversational interface for user interaction.
• ChromaDB: A database for managing vector data.
• DuckDNS: A dynamic DNS service for easy access.
• FastAPI: A modern web framework for building APIs.
• LangChain: A framework for building applications with LLMs.
• Local LLM: Run models locally for privacy and speed.
• Nginx: A high-performance web server.
• Ollama: A tool for managing and deploying LLMs.
• RAG: Combines retrieval and generation for enhanced performance.
• React: A JavaScript library for building user interfaces.
• Vite: A build tool that provides a fast development environment.

Installation

To get started with Deploy Safe, follow these steps to install the necessary components.

Prerequisites

Ensure you have the following installed:

• Python 3.7 or higher
• Node.js
• Docker (optional, for containerized deployment)

Steps

1. Clone the Repository

   git clone https://github.com/duribeiro04/deploy-safe.git
   cd deploy-safe

2. Install Python Dependencies

   pip install -r requirements.txt

3. Install Node.js Dependencies

   cd frontend
   npm install

4. Set Up Environment Variables

   Create a .env file in the root directory and configure your settings.

5. Run the Application

   For development, use:

   uvicorn main:app --reload

   For production, use Docker:

   docker-compose up -d

Usage

After installation, you can start using Deploy Safe.

1. Access the Application

   Open your browser and navigate to http://localhost:8000 to access the interface.

2. Interact with the Chatbot

   Use the chatbot interface to input queries and receive responses.

3. Explore Semantic Search

   Test the semantic search functionality by entering Chinese text.

Configuration

Customize your application by modifying the configuration files.

Environment Variables

Here are some key environment variables you can set:

• DATABASE_URL: URL for your ChromaDB instance.
• DUCKDNS_TOKEN: Your DuckDNS token for domain access.
• NGINX_CONFIG: Path to your Nginx configuration file.

Nginx Configuration

Ensure that your Nginx configuration is set up to serve the application correctly. An example configuration is provided in the nginx directory.

Deployment

For deploying your application, you can choose between local and cloud hosting options.

Local Deployment

Follow the installation steps to run the application locally. Use Docker for easier management.

Cloud Deployment

You can deploy Deploy Safe on platforms like AWS, DigitalOcean, or Heroku.

1. Create a Server Instance
2. Install Dependencies
3. Clone the Repository
4. Run the Application

For detailed steps, refer to the documentation of your chosen cloud provider.

Contributing

Contributions are welcome! Please follow these steps:

1. Fork the repository.
2. Create a new branch (git checkout -b feature/YourFeature).
3. Make your changes.
4. Commit your changes (git commit -m 'Add some feature').
5. Push to the branch (git push origin feature/YourFeature).
6. Open a pull request.

License

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

Contact

For questions or suggestions, please reach out:

• GitHub: duribeiro04
• Email: your.email@example.com

For the latest releases, visit here. Download and execute the necessary files to get started with Deploy Safe.