The 22nd of May 2024, Daniel and Jettro will visit DeFabrique in Utrecht for the annual Teqnation conference. In the morning, we will deliver our workshop The Art of Questions: Creating a Semantic Search-Based Question-Answering System with LLMs. This workshop makes use of our RAG4j/p project to teach a group of enthusiasts about the different aspects of Retrieval Augmented Generation. We will use the project to show the different parts of a RAG system. Time is short, still we want participants to experience the essential parts of a RAG system. As the project is small, little time is lost learning it.

People attending the workshop will start from the beginning and perform all the different steps needed to build and evaluate a RAG system. To give you an idea of the different steps, here is a list of the different steps:

  • Work with text; use a splitter to split the text into sentences or other chunks.
  • Use an embedder to create vectors from the chunks
  • Use a retriever to find the most relevant chunks for a given question
  • Use a generator to generate an answer based on the chunks and the question
  • Use a quality component to determine the quality of the generated text and the relevance of the retrieved chunks

We hope to see you in DeFabrique and that you will enjoy the workshop. If you have any questions, feel free to contact us.

22nd of May 2024 11:00 - 12:45 Room KALVERMELK 2C

Preparing for the workshop

First choose your programming language (Java or Python). Then follow the instructions below:

Java: RAG4j

Python: RAG4p

The repositories contain a README.md file with instructions on how to setup your environment.

 

Workshop Image