video1.0<iframe src="https://www.loom.com/embed/476c1addac204830bdbc60292ad52d4e" frameborder="0" width="1728" height="1296" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>12961728Loomhttps://www.loom.com12961728https://cdn.loom.com/sessions/thumbnails/476c1addac204830bdbc60292ad52d4e-00001.gif300.6In this video, I give a high level walkthrough of building your first RAG/RAQA application: - Split docs into chunks - create embeddings for each chunks using an openai embedding model - Store embeddings in a local vector database - Wrap the vector store in a Retriever - Take the user query and compute cosine similarity between our question and stuff in vector store - Setup Visiblity and Eval using Wandb so that we can inspect what went wrong - Using GPT4 as a custom evaluator for our RAG application Github Notebook Link: https://github.com/rajkstats/AIE2/blob/main/Week%202/Day%201/Pythonic%20RAG%20Assignment_rk.ipynb