Practical RAG - Choosing the Right Embedding Model, Chunking Strategy, and More
ฝัง
- เผยแพร่เมื่อ 8 พ.ย. 2023
- In this informative talk, Frank Liu, Head of AI & ML at Zilliz, dives into the practical aspects of Retrieval Augmented Generation (RAG).
He discusses the importance of choosing the right embedding model, chunking strategy, and more when building a RAG application. Frank also provides insights into vector databases and their role in handling unstructured data. Additionally, he offers a sneak peek into the future of vector databases and their potential applications beyond large language models. If you're interested in AI tools and want to learn best practices from industry experts, don't miss this talk! #AItools #RAG #VectorDatabases #ArtificialIntelligence #MachineLearning - วิทยาศาสตร์และเทคโนโลยี
man if you had a youtube channel where you made short videos like this teaching stuff. I would religiously follow you lol. this was more informative than all the video I have seen so far in the sense that it answered some of the most annoying questions I had (annoying as in couldn't find answer to them) and more.
Thank you for sharing these best practice insights.
Superb concise view thanks Frank
great talk, appreciated!: Question: when it comes to choosing the vectorstore (around @11:00): are we talking about size per "client database" or for your "complete (multiple, combined) database" ? What if our clients on average have a KB of 10 GB ?
Very nice talk
A good explanation to prepare for building a rag model, could you make a following deep dive tutorial to thoroughly understand the process and intricacies involved? Many thx.