From RAG to RAG Fusion to RAPTOR An AI Development Journey

In this video we will explore a development path from RAG (retrieval augmented generation) to RAG Fusion (RAG + Reciprocal Ranking Fusion) to RAPTOR (Recursive Abstractive Processing for Tree-Organized Retrieval) using Ollama and some open source language models
The repository for Packet Raptor can be found here:
github.com/automateyournetwor...
The RAPTOR paper:
arxiv.org/html/2401.18059v1
The RAPTOR cookbook:
github.com/langchain-ai/langc...
Ollama:
ollama.com/
github.com/ollama/ollama
Open WebUI:
github.com/open-webui/open-webui
Follow me on X:
/ john_capobianco

[@ademiry]

Very nice and easy to understand video.. I shared it on my LI

[@yt-caio]

This was a master class. Thanks for sharing.

[@johncapobianco2527]

I really appreciate that thank you

[@niclasthegoat]

you are an absolute legend! found your video 10 minutes ago and the value pluse your way of transporting your knowladge is amazing!

[@johncapobianco2527]

that is very kind and I'm really glad it helped out just in time!

[@carolinacamachogarcia5158]

I love your videos, John. Thank you for sharing

[@johncapobianco2527]

Thank you so much :D

[@sunandong]

thanks a lot , the vedio helps me very much

[@arifzrg7875]

Hi there, how about doing a video on xRAG it is using a compression of embeddings, would really be helpful if you can explain of its workings. Thnx in advance

[@andydataguy]

Just discovered your channel. Been wanting to learn more about a network engineer's perspective on RAPTOR!! Thanks for making my dream come true today 🙏🏾💜

[@johncapobianco2527]

🥰

[@andydataguy]

@@johncapobianco2527 I've watched a lot of videos on the subject. I'd love to see your take on RAGAS for evaluation. I''m sure there's an exciting way to merge that with RAPTOR for truly robust systems

[@SDAravind]

Any thoughts, on RAGGraph vs RAPTOR? Using qdrant db as private db?

[@prasad_yt]

At the start - for simple RAG You mentioned that LLM is involved in retrieval process , is that a correct statement?

[@johncapobianco2527]

Correct with the ConversationalRetreivalChain.from_llm()

[@NhatNguyen-bq6jj]

Can you summarize large documents with this technique?

[@johncapobianco2527]

Yes the point of RAPTOR is to move from RAG (small documents) to RAPTOR (larger documents)

[@NhatNguyen-bq6jj]

@@johncapobianco2527 Can you make video show how to use RAG to summarize multi documents? Thank you!