RAG-PDF-Search is a Retrieval-Augmented Generation (RAG) application built with Streamlit, designed for intelligent searching within PDF documents. By leveraging Ollama and DeepSeek R1, it provides highly relevant, context-aware responses based on document content. The application also supports Arabic text processing and delivers concise answers to user queries.
- 📂 Upload and search within PDFs with ease.
- 🧠 AI-powered responses using DeepSeek R1 via Ollama.
- 🔍 Efficient document retrieval powered by FAISS-based vector search.
- 📚 Advanced text chunking for enhanced semantic understanding.
- 🌍 Multilingual and Arabic text support for diverse use cases.
- ⚡ Fast and interactive UI built with Streamlit.
Ensure you have Python 3.8+ installed and all required dependencies.
git clone https://github.com/NASO7Y/RAG-PDF-Search.git
cd RAG-PDF-Searchpip install -r requirements.txtstreamlit run RAG.py- Python (Core development language)
- Streamlit (User interface framework)
- LangChain (AI-powered retrieval and processing)
- Ollama & DeepSeek R1 (Natural language processing models)
- FAISS (Fast vector-based search)
- HuggingFace Embeddings (Semantic text embeddings)
- PDFPlumber (PDF document processing)
- Upload a PDF file via the Streamlit interface.
- The application extracts, processes, and embeds the text using HuggingFace embeddings.
- Queries are matched to relevant document segments using FAISS-based retrieval.
- Ollama & DeepSeek R1 generate a precise, context-aware response.
- The results are displayed in a user-friendly Streamlit UI.
We welcome all contributions! Feel free to fork the repository, submit issues, or create pull requests.
For any questions or feedback, feel free to reach out:
- GitHub: NASO7Y
- Email: [email protected]
- LinkedIn: Ahmed Noshy
⭐ If you find this project helpful, consider giving it a star is support😂🌹