search

FuturePathAI / Learn-AI-Engineering

Code, notebooks, and other material for FuturePath AI's training course on Generative AI
MIT License
8 stars 6 forks source link

RAG 101 with Langchain, Unstructured and Llama Index #47

Open NirantK opened 5 months ago

NirantK commented 5 months ago
  1. Problem Motivation: Directory of pdfs which are more than 100K tokens
  2. Getting started with Langchain and Unstructured
  3. Grounding: Connecting Search to LLM to get best answers

Notebook name: Week2-Lecture2-Langchain-Unstructured.ipynb

RheagalFire commented 5 months ago

@NirantK We already have one notebook for this here . Is this supposed to be an extension to this notebook , if yes than can we have a more granular breakdown of what needs to be covered here?

NirantK commented 5 months ago
  1. Needs to change from a single web page to a directory of pdfs
  2. Use unstructured or Azure Document Intelligence or Llama Parser → We use BS4 right now