This project is a part of my self-development Retrieval-Augmented Generation (RAG) application that allows users to ask questions about the content of a PDF files placed in folder. The app uses techniques to provide accurate answers based on the document's content. The application leverages Ollama, Llama 3-8B, LangChain, and FAISS for its operations.
- Ask Questions About PDFs: Simply place a PDF file in the
datafolder or upload from Web UI and start asking questions about its content. - LLM Models: Utilizes Ollama and Llama 3-8B for generating responses.
- Efficient Document Retrieval: Uses LangChain and FAISS for efficient document retrieval and processing (PDF Only).
- Duplicate Handling: The app checks the vector database for duplicates and avoids adding them if they already exist.
- Python 3.8 or higher
- Required Python packages (see
requirements.txt) - Ollama Installation with Llama 3 installed
- Clone the repository:
git clone https://github.com/emre570/simple-rag-app.git cd simple-rag-app - Install the required packages:
pip install -r requirements.txt
- Download Ollama and install LLM using Ollama:
ollama pull llama3
-
Place your PDF file(s) in the
datafolder or upload it from Web UI. -
Run the application:
streamlit run [path_to_app_folder]/webui.py
Alternatively, you can open
webui.batfile. -
Enter your questions when prompted. Type 'exit' to quit the application.
- Web UI: A web-based user interface for easier interaction.
- Conversation Memory: The app will remember previous interactions during runtime for better context handling.
- Support for Multiple Document Types: Extend functionality to work with Powerpoint slides, markdown files, text files, and more.
Contributions are welcome! Please fork the repository and create a pull request with your changes. For changes or see any mistakes, please open an issue first to discuss what you would like to change.
For any questions or suggestions, please open an issue in the repository.