Search icon CANCEL
Subscription
0
Cart icon
Your Cart (0 item)
Close icon
You have no products in your basket yet
Arrow left icon
Explore Products
Best Sellers
New Releases
Books
Videos
Audiobooks
Learning Hub
Conferences
Free Learning
Arrow right icon
Arrow up icon
GO TO TOP
RAG-Driven Generative AI

You're reading from   RAG-Driven Generative AI Build custom retrieval augmented generation pipelines with LlamaIndex, Deep Lake, and Pinecone

Arrow left icon
Product type Paperback
Published in Sep 2024
Publisher Packt
ISBN-13 9781836200918
Length 334 pages
Edition 1st Edition
Languages
Tools
Concepts
Arrow right icon
Author (1):
Arrow left icon
Denis Rothman Denis Rothman
Author Profile Icon Denis Rothman
Denis Rothman
Arrow right icon
View More author details
Toc

Table of Contents (14) Chapters Close

Preface 1. Why Retrieval Augmented Generation? 2. RAG Embedding Vector Stores with Deep Lake and OpenAI FREE CHAPTER 3. Building Index-Based RAG with LlamaIndex, Deep Lake, and OpenAI 4. Multimodal Modular RAG for Drone Technology 5. Boosting RAG Performance with Expert Human Feedback 6. Scaling RAG Bank Customer Data with Pinecone 7. Building Scalable Knowledge-Graph-Based RAG with Wikipedia API and LlamaIndex 8. Dynamic RAG with Chroma and Hugging Face Llama 9. Empowering AI Models: Fine-Tuning RAG Data and Human Feedback 10. RAG for Video Stock Production with Pinecone and OpenAI 11. Other Books You May Enjoy
12. Index
Appendix

1. Preparing the dataset for fine-tuning

Fine-tuning an OpenAI model requires careful preparation; otherwise, the fine-tuning job will fail. In this section, we will carry out the following steps:

  1. Download the dataset from Hugging Face and prepare it by processing its columns.
  2. Stream the dataset to a JSON file in JSONL format.

The program begins by downloading the dataset.

1.1. Downloading and visualizing the dataset

We will download the SciQ dataset we embedded in Chapter 8. As we saw, embedding thousands of documents takes time and resources. In this section, we will download the dataset, but this time, we will not embed it. We will let the OpenAI model handle that for us while fine-tuning the data.

The program downloads the same Hugging Face dataset as in Chapter 8 and filters the training portion of the dataset to include only non-empty records with the correct answer and support text to explain the answer to the questions:

# Import required...
lock icon The rest of the chapter is locked
Register for a free Packt account to unlock a world of extra content!
A free Packt account unlocks extra newsletters, articles, discounted offers, and much more. Start advancing your knowledge today.
Unlock this book and the full library FREE for 7 days
Get unlimited access to 7000+ expert-authored eBooks and videos courses covering every tech area you can think of
Renews at $19.99/month. Cancel anytime