README.md

metadata

title: CSRC Car Manual RAG System
emoji: 🚗
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 6.0.0
app_file: app.py
pinned: false
license: mit

CSRC Car Manual RAG System

An intelligent RAG (Retrieval-Augmented Generation) system for querying car manual documents using OpenAI and vector stores.

🚀 Features

• RAG-based Q&A: Ask questions about car manual content
• Vector Store: Fast and accurate document retrieval
• Knowledge Graph: Visualize document relationships
• Personalized Learning: Adaptive learning paths (optional)
• Scenario Contextualization: Context-aware responses (optional)

📋 Setup Instructions

1. Clone or Upload to Hugging Face Spaces

• Option A: Create a new Space on Hugging Face and upload files
• Option B: Connect your GitHub repository to Spaces

2. Set Environment Variables (Secrets)

Go to Settings > Secrets in your Space and add:

OPENAI_API_KEY=your-openai-api-key-here

⚠️ Important: Never commit API keys to the repository. Always use Spaces Secrets.

3. Upload PDF Files

Ensure your PDF files are in the car_manual/ directory:

car_manual/
├── Function of Active Distance Assist DISTRONIC.pdf
├── Function of Active Lane Change Assist.pdf
├── Function of Active Steering Assist.pdf
└── Function of Active Stop-and-Go Assist.pdf

4. Wait for Build

Spaces will automatically:

• Install dependencies from requirements.txt
• Run app.py
• Start the Gradio interface

📁 Project Structure

.
├── app.py                 # Hugging Face Spaces entry point
├── main.py               # Local development entry point
├── requirements.txt      # Python dependencies
├── src/                  # Core modules
├── modules/              # Feature modules
├── car_manual/           # PDF files directory
├── config/               # Configuration files
└── output/               # Output directory (auto-created)

🔧 Configuration

Required

• OPENAI_API_KEY: Your OpenAI API key (set in Spaces Secrets)

Optional

• PDF Files: Place in car_manual/ directory
• Vector Store: Automatically created on first run

📖 Usage

1. Wait for the Space to build (check the logs)
2. Open the Gradio interface
3. Enter your question in the input field
4. Get answers with source citations

🐛 Troubleshooting

Error: OPENAI_API_KEY not found

• Go to Settings > Secrets
• Add OPENAI_API_KEY with your actual API key
• Restart the Space

Error: No PDF files found

• Ensure PDF files are in the car_manual/ directory
• Check file permissions
• Verify file names (case-sensitive)

Build Fails

• Check the logs for error messages
• Verify requirements.txt is correct
• Ensure all Python dependencies are compatible

📝 Notes

• Vector store is created automatically on first run
• Vector store ID is saved in config/vector_store_config.json
• First initialization may take time (uploading PDFs to OpenAI)

🔗 Links

• OpenAI API Keys
• Hugging Face Spaces Documentation

📄 License

MIT License