Welcome to the RAG Workshop! This repository is your starting point to dive into the world of Retrieval-Augmented Generation (RAG). Whether you're curious about how RAG works, how to optimize it, or how to build cool AI-powered chat apps, you're in the right place!
Please find the document that will serve as the content for the RAG app we are developing here.
In this workshop, you will:
- Build a RAG system from scratch: Learn how to create a Retrieval-Augmented Generation (RAG) solution, covering everything from prompting to the basics of setting up a vector database.
- Master the art of prompting: Discover how to craft prompts that effectively guide your RAG models for optimal results.
- Work on a real-world project: Bring it all together by building a Q&A chat app for study material, using Streamlit.
While the workshop focuses on a Q&A application for educational content, the skills you'll gain are versatile and applicable to various scenarios, such as improving productivity of customer support by enabling querying policies or analysing customer reviews. RAG empowers language models by enriching them with domain-specific knowledge to tackle diverse tasks.
- Create a github account if you dont have one
- Fork this repo
- Open it in codespaces to develop!
We prepared some video tutorials so you can go fast to the code :)
Here's what you'll find in this repository:
- 📓 notebooks: Step-by-step guides to get you up and running with RAG basics.
- 💬 chat_solution: The final chat application you'll build after completing the notebooks.
- 📂 data: All the datasets you need to work through the exercises. You can also use your own PDFs for a more personal touch!
And tons of other configurations files that you do not need to worry about now.
Go to notebook 1 to get started.
