Master RAG on Vertex AI with Vector Search and Gemini Pro
ฝัง
- เผยแพร่เมื่อ 29 มิ.ย. 2024
- Are you ready to take your question-answering systems to the next level? In this tutorial, we'll dive into integrating Retrieval Augmented Generation (RAG) with Google Cloud's Vertex AI Vector Search and the powerful Gemini language model.
You can access the complete code at gist.github.com/janakiramm/55... (Vector Search)
and
gist.github.com/janakiramm/7d... (RAG)
What you'll learn:
Understanding RAG: How RAG combines retrieval and generative techniques for superior question answering.
Setting up Vertex AI Vector Search: Create and configure your vector search index for efficient document storage and retrieval.
Harnessing Gemini: Leverage Gemini's language capabilities to enhance RAG's answer generation.
Step-by-step Implementation: Follow along as we build a RAG-powered question-answering system on Vertex AI.
Tips and Best Practices: Get insights for optimizing your RAG implementation.
Chapters:
00:00 Introduction
00:54 Overview of RAG
07:05 Configuring and Deploying Vector Search Index Endpoint
18:10 RAG with Gemini
LinkedIn: / janakiramm
#RAG #QuestionAnswering #GoogleCloud #VertexAI #VectorSearch #Gemini #subscribe #genai #tutorial - วิทยาศาสตร์และเทคโนโลยี
Thanks, this is tremendously helpful
One point to note - you need to upload the embed file, not the sentence file -> upload_file(bucket_name,embed_file_path)
Best tutorial. Big thanks for your shared.
Excelent video! Thanks for sharing the code too.
Glad it was helpful!
Great Video, thank you soo much........
Thanks for the tutorial!
Instead of going through the ids in the json file to fetch the sentences, is it possible to integrate those directly as metadata in the index?
Great video! What is the difference between Vertex Search service VS Vector Search for RAG application? which one is better in terms of handling better retrieval of relevant documents for RAG application where we deal with 100+ PDF documents? Can you share some insights?
Thanks for the tutorial. I am bit confused which file to be uploaded to bucket. sentence file or embedding file
Great!
Great job! Thanks a lot. What’s the difference between this approach and using langchain?
Excellent video - can u please do same with Langchain with retrieval
Great Video. One question, I noticed you used a different model (gecko) to Gemini Pro for the embeddings. Is this ok to do? I assumed the models needed to be the same for both training and inference? Thanks again
Text embedding models are independent of LLMs. You only have to ensure that the same embedding model is used for indexing the documents and the query. This is critical to retrieving the context based on the similarity.
the code link u have shared is incomplete, load_file is missing and other few stuffs,
Can you please do a video on "How to use the same in Langchain with retrieval"
+1
Nice. Are you ok to share the colab notebook?
Yes, sure. Please check the description. I have added the links.
Thanks for sharing knowledge.
Can you share the notebook
Please check the description. I have added the links.
why always python is there any way to use js?
Possible to share the notebook?
The code is available at gist.github.com/janakiramm/55d2d8ec5d14dd45c7e9127d81cdafcd and gist.github.com/janakiramm/7dd73e83c92a0de0c683ed27072cdde2