RAG (Retrieval-Augmented Generation) System

A production-ready RAG system with Qt-based UI for document indexing and intelligent question answering.

🚀 Features

• Multi-Provider LLM Support: Gemini, OpenAI, Claude
• Advanced Document Processing: Multiple chunking strategies (adaptive, semantic, fixed, sentence-based)
• Intelligent Retrieval: Vector search with BM25 reranking
• Real-time UI: Qt-based desktop application with progress tracking
• Asynchronous Processing: Background document ingestion
• Configurable Pipeline: Modular architecture with dependency injection

📋 Prerequisites

• Python 3.8+
• Qt/PySide6
• ChromaDB for vector storage

🔧 Installation

1. Clone the repository

git clone [your-repo-url]
cd Ragproject

2. Create virtual environment

python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies

pip install -r requirements.txt

4. Set up environment variables

# Copy the sample environment file
cp .env.sample .env

# Edit .env and add your API keys
# At minimum, you need to set:
# GEMINI_API_KEY=your_actual_api_key_here

🔑 API Keys Setup

Gemini (Required)

1. Go to Google AI Studio
2. Create a new API key
3. Add to .env: GEMINI_API_KEY=your_key_here

OpenAI (Optional)

1. Visit OpenAI Platform
2. Create an API key
3. Add to .env: OPENAI_API_KEY=your_key_here

Claude (Optional)

1. Go to Anthropic Console
2. Generate an API key
3. Add to .env: ANTHROPIC_API_KEY=your_key_here

🎯 Quick Start

Cross-Platform Launch (Recommended)

Windows:

start_system.bat

Linux/Mac:

chmod +x start_system.sh
./start_system.sh

Or directly with Python:

python main.py

Advanced Options

Run server only:

python main.py --server

Run UI only (server must be running):

python main.py --ui

Specify custom port:

python main.py --port 8000

📁 Project Structure

Ragproject/
├── main.py                # Main entry point
├── run_server.py         # Server launcher
├── qt_app.py            # Qt UI application
├── rag/                 # Core RAG system
│   ├── core/           # Core components
│   ├── pipeline/       # Pipeline steps
│   └── chunkers/       # Document chunking
├── server/             # FastAPI server
│   ├── routers/       # API endpoints
│   └── dependencies.py # Dependency injection
├── ui/                 # Qt UI components
│   ├── chat_widget.py # Main chat interface
│   └── options/       # Settings tabs
├── config/            # Configuration files
│   └── config.yaml   # Server configuration
└── chroma_db/        # Vector database storage

⚙️ Configuration

Retrieval Settings

• retrieve_k: Number of documents to retrieve from vector store (default: 20)
• rerank_k: Number of documents after reranking (default: 5)
• max_context_chars: Maximum context size in characters (default: 12000)

Chunking Strategies

• Adaptive: Smart chunking based on document structure
• Semantic: Meaning-based chunking using embeddings
• Fixed: Fixed-size chunks with overlap
• Sentence: Sentence-based chunking

🛠️ Development

Code Style

# Format code
black .

# Check linting
pylint rag/

TODO

제미니 API에 최적화된 부분 LLM Client, 래핑해서 모두에게 맞게 하기 평가 단계 추가

AdaptiveChunker 개선: 단순 휴리스틱을 넘어, langdetect와 같은 라이브러리를 사용하여 콘텐츠의 언어와 유형(산문, 코드, 표 등)을 명시적으로 탐지하는 단계를 추가. 이를 통해 각 유형에 맞는 전문 청커(e.g., CodeChunker)로 작업을 위임하는 **전략 라우팅(Strategy Routing)**을 구현

EmbedderManager 정책 고도화: koRatio 대신, 토큰화된 결과물을 분석하여 한국어 형태소의 비율을 측정하는 등 더 정교한 언어 감지 로직을 도입 또한, 사용자가 특정 문서 그룹에 대해 임베딩 프로필을 수동으로 오버라이드할 수 있는 기능을 제공하여 정책의 실패를 보완

BM25Reranker 교체: **형태소 분석기(e.g., mecab-ko, okt)**를 사용하여 텍스트를 토큰화하는 BM25Reranker를 구현. rank_bm25 라이브러리는 커스텀 토크나이저 함수를 지원하므로 쉽게 통합할 수 있음

deleteByDoc 최적화: collection.get()과 collection.delete(ids=...)의 2단계 프로세스를 collection.delete(where={"docId": docId})의 단일 API 호출로 변경

프로덕션 배포 스크립트 분리: run_server.py (개발용, reload=True)와 별도로, Gunicorn과 uvicorn 워커를 사용하는 프로덕션용 실행 스크립트(run_prod.sh 등)와 Dockerfile을 작성하여 배포 환경을 표준화

현재 모든 파이프라인을 Unit화해서 단계를 노드로 해서 그래프화 할것

RAG 기능 끄고 켜기 추가해서, RAG 없이 단순 채팅용 LLM으로도 쓸수있게

📝 License

[Your License]

🤝 Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

⚠️ Important Security Notes

• NEVER commit your .env file with actual API keys
• Always use .env.sample as a template
• Keep your API keys secure and rotate them regularly
• Review .gitignore before committing

🐛 Troubleshooting

Server won't start

• Check if port 7001 is already in use
• Verify all dependencies are installed
• Ensure .env file exists with valid API keys

Document ingestion fails

• Check file permissions
• Verify supported file formats (.txt, .md, .pdf)
• Check available disk space for vector database

UI not responding

• Restart both server and Qt application
• Check server logs for errors
• Verify network connectivity to localhost:7001

📧 Contact

[Your Contact Information]