ฝัง
- เผยแพร่เมื่อ 5 ก.พ. 2025
- LLMs are a powerful new platform, but they are not always trained on data that is relevant for our tasks. This is where retrieval augmented generation (or RAG) comes in: RAG is a general methodology for connecting LLMs with external data sources such as private or recent data. It allows LLMs to use external data in generation of their output. This video series will build up an understanding of RAG from scratch, starting with the basics of indexing, retrieval, and generation. It will build up to more advanced techniques to address edge cases or challenges in RAG.
Code:
github.com/lan...
Slides:
docs.google.co...
Having more, shorter videos is really helpful and makes the content far easier to consume and learn from. Thanks for teaching us!
Please keep making tutorial playlists with shorter videos.
This series is gold! The short format is great. Thanks so much for making it publicly available
Great tutorial. 2 small fixes to get the code running. (1) add bs4 to ! pip install (2) and the os.environ['OPENAI_API_KEY'] = 'Your key' to import os. Have fun!
So I must pay in order to be able to follow up with this tuto ?
short videos are a great idea , it works well and the explanations are as well good
Great walkthrough. Thank you for taking the time to put this together.
Would love a video series going over using langchain in combination with your own local SQL data.
🎯 Key points for quick navigation:
00:03 *📹 The "RAG from Scratch" series will cover basic principles and advanced topics for building LLM applications with LangChain.*
00:15 *🔒 LLMs haven't seen all data, including private or recent data, due to limited pre-training runs.*
00:44 *📊 LLMs have context windows that are increasing in size, representing dozens to hundreds of pages of information.*
01:10 *💻 Retrieval-Augmented Generation (RAG) is a popular paradigm for connecting LLMs to external data, involving three stages: indexing, retrieval, and generation.*
02:06 *📝 Future videos will explore methods and tricks for RAG's three basic components in detail.*
Made with HARPA AI
Very nice, thanks! It would be great to see a RAG tutorial that was totally off the internet, well, except for installation and downloading. I can't find any such demo's; there's always an API of some sort. In a way, I get it, but still it would be nice to see a RAG with WiFi off and no internet ... just once. Thanks as I've learned a bit more.
(This is Lance from the video.) I did one on local RAG. It shows how to set up local LLM and embedding model w/ Ollama and Nomic: th-cam.com/video/E2shqsYwxck/w-d-xo.html
Looks promising! I will keep watching the playlist
Thank you. This series is super helpful!
Thanks so much for the hands on tutorial! Very nicely made.
Great playlist!
Nice walkthrough. thanks!
Can you use a higher resolution for your videos? At least 1080p for anything with text.
(Lance is Lance from the video.) Yes, will do. It was a problem w/ Loom. Apologies.
Great video :) thank you
Hoping that too have production ready checklists, Hoping to do the same with JavaScript
Hi,
Great tutorial series here. A quick question as I am not able to find the right documentation for this. Can you tell me how rag_chain = ( | ).invoke snippet works. Per my understanding we are piping one result to another as we would with grep. is it similar?
Guest We're too early for the Notebook to be ready. #timeofcomment
ok np
Hi, how are you bypassing rate limit error of openAi
unfortunately, under 720P resolution, some text in the slides are very difficult to see clearly. I will be much better, if we can get 1080p or even higher resolution.
There is a link to the slides in the description.
@@nellatara many thanks to your help
Could you provide the code demo? Thanks!
Just wondering, where did the part 15 reranking go? I can’t find the video.
So RAG builds the prompt and this can scale with context window size?
Does anyone know in which tool those block diagrams are made?
Can RAG helpful to text-image-generation?
great start. but the code link is broken. 404 error.
Great video, but please fix the code link.
How can I run those Jupiter notebook locally, i've clone them now what?
anything beyond "frozen" rag?
why 720p?
0:50
code URL == 404
Yes, It would be great to have access
gemini 1.5