This video presents a basic way to use “retrieval augmented generation” (RAG) with your own Google Drive as a knowledge source. Plug-and-play, it uses Google Colab, which can be accessed for free, and Mistral AI open source models, with API calls that should only cost a fraction of a cent per run. The app could run with a single click, but I left distinct cells and sections as going through these steps helps construct an understanding of the underlying technology, showing that RAG involves:
📦 Splitting the source in chunks, calculating their embeddings, and storing them in a vector store 🔍 Retrieving the most relevant ones based on the user query 📝 Using them to generate the final response Much more advanced strategies, functions and layers could be added to this app to increase its performance, UX and aesthetics - but its simplicity makes it a good starting point for people (including students) interested in the inner workings of artificial intelligence. 👉 You can try the app for yourself here: https://bit.ly/mistral_RAG To do something similar with online webpages rather than PDFs on your Drive, you can follow this great guide created by Mistral, which was my inspiration here: https://docs.mistral.ai/guides/basic-RAG/
0 Comments
Your comment will be posted after it is approved.
Leave a Reply. |
|