leemark / llamaindex_playbook

A collection of basic Python examples utilizing LlamaIndex to showcase various chat interfaces and Retrieval-Augmented Generation (RAG) strategies

LlamaIndex RAG Playbook

Introduction

This repository contains a collection of basic Python examples utilizing LlamaIndex to showcase various chat interfaces and Retrieval-Augmented Generation (RAG) strategies. Each example is designed to be self-contained and demonstrates a unique aspect of working with RAG and chatbot interfaces.

Requirements

• Python 3.8+
• LlamaIndex library
• Additional requirements are listed in the requirements.txt file.

Installation

To set up your environment to run these examples, follow these steps:

git clone https://github.com/leemark/llamaindex_ex.git
cd llamaindex_ex
pip install -r requirements.txt

Usage

To run an example, navigate to its directory and execute the Python script:

cd path/to/examples
python hello.py

Examples

• hello.py: Demonstrates a basic "Hello, World!" example with an index.
• hello_persist.py: Showcases how to create a persistent RAG index.
• local_models_ollama.py: Provides an example of using local models with LlamaIndex and ollama.
• streamlit_interface.py: A simple Streamlit app interface for interacting with the model.
• rag_web_page.py: An example of augmenting a single web page with RAG.
• rag_web_search_brave.py: An example of augmenting brave api web search results with RAG.

Contributing

We welcome contributions! If you'd like to improve an example or add a new one, please open a pull request with your proposed changes.

License

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

Contact

If you have any questions or want to discuss the examples further, feel free to reach out.