book

Data Analysis with LLMs

Name: Data Analysis with LLMs
Author: Immanuel Trummer
ISBN: 9781633437647

by Immanuel Trummer

April 2025

Intermediate to advanced

232 pages

6h 56m

English

Manning Publications

Read now

Unlock full access

Data Analysis with LLMs
copyright
contents
dedication
preface
acknowledgments
about this book
about the author
about the cover
Part 1 Introducing language models

1 Analyzing data with large language models
1.1 What can language models do?1.2 What you will learn1.3 How to use language models1.3.1 Prompting1.3.2 Example prompt1.3.3 Interfaces1.4 Using language models for data analysis1.4.1 Using language models directly on data1.4.2 Data analysis via external tools1.5 Minimizing costs1.5.1 Picking the best model1.5.2 Optimally configuring models1.5.3 Prompt engineering1.6 Advanced software frameworks and agents
2 Chatting with ChatGPT
2.1 Accessing the web interface2.2 Making introductions2.3 Processing text with ChatGPT2.4 Processing tables with ChatGPT2.4.1 Processing tables in the web interface2.4.2 Processing tables on your platform2.6 References
Part 2 Data analysis with language models
3 The OpenAI Python library
3.1 Prerequisites3.2 Installing OpenAI’s Python library3.3 Listing available models3.4 Chat completion3.5 Customizing model behavior3.5.1 Configuring termination conditions3.5.2 Configuring output generation3.5.3 Configuring randomization3.5.4 Customization example3.5.5 Further parameters
4 Analyzing text data
4.1 Preliminaries4.2 Classification4.2.1 Overview4.2.2 Creating prompts4.2.3 Calling the model4.2.4 End-to-end classification code4.2.5 Classifying documents4.2.6 Running the code4.2.7 Trying out variants4.3 Text extraction4.3.1 Overview4.3.2 Generating prompts4.3.3 Postprocessing4.3.4 End-to-end extraction code4.3.5 Trying it out4.4 Clustering4.4.1 Overview4.4.2 Calculating embeddings4.4.3 Clustering vectors4.4.4 End-to-end code for text clustering4.4.5 Trying it out4.4.6 Other use cases for embedding vectors4.6 References
5 Analyzing structured data
5.1 Chapter outline5.2 A natural language query interface for analyzing game sales5.2.1 Setting up an SQLite database5.2.2 SQL basics5.2.3 Overview5.2.4 Generating prompts for text-to-SQL translation5.2.5 Complete code5.2.6 Trying it out5.3 A general natural language query interface5.3.1 Executing queries5.3.2 Extracting the database structure5.3.3 Complete code5.3.4 Trying it out5.4 A natural language query interface for graph data5.4.1 What is graph data?5.4.2 Setting up a Neo4j database5.4.3 The Cypher query language5.4.4 Translating questions to Cypher queries5.4.5 Generating prompts5.4.6 Complete code5.4.7 Trying it out
6 Analyzing images and videos
6.1 Setup6.2 Answering questions about images6.2.1 Specifying multimodal input6.2.2 Code discussion6.2.3 Trying it out6.3 Tagging people in images6.3.1 Overview6.3.2 Encoding locally stored images6.3.3 Sending locally stored images to OpenAI6.3.4 The end-to-end implementation6.3.5 Trying it out6.4 Generating titles for videos6.4.1 Overview6.4.2 Encoding video frames6.4.3 The end-to-end implementation6.4.4 Trying it out
7 Analyzing audio data
7.1 Preliminaries7.2 Transcribing audio files7.2.1 Transcribing speech7.2.2 End-to-end code7.2.3 Trying it out7.3 Querying relational data via voice7.3.1 Preliminaries7.3.2 Overview7.3.3 Recording audio7.3.4 End-to-end code7.3.5 Trying it out7.4 Speech-to-speech translation7.4.1 Overview7.4.2 Generating speech7.4.3 End-to-end code7.4.4 Trying it out
Part 3 Advanced topics
8 GPT alternatives
8.1 Anthropic8.1.1 Chatting with Claude8.1.2 Python library8.2 Cohere8.2.1 Chatting with Command R+8.2.2 Python library8.3 Google8.3.1 Chatting with Gemini8.3.2 The Python library8.4 Hugging Face8.4.1 Web platform8.4.2 Python library8.6 References
9 Optimizing cost and quality
9.1 Example scenario9.2 Untuned classifier9.3 Model tuning9.4 Model selection9.5 Prompt engineering9.6 Tunable classifier9.7 Fine-tuning9.8 Generating training data9.9 Starting a fine-tuning job9.10 Using the fine-tuned model
10 Software frameworks
10.1 LangChain10.2 Classifying reviews with LangChain10.2.1 Overview10.2.2 Creating a classification chain10.2.3 Putting it together10.2.4 Trying it out10.3 Agents: Putting the large language model into the driver’s seat10.4 Building an agent for data analysis10.4.1 Overview10.4.2 Creating an agent with LangChain10.4.3 Complete code for data-analysis agent10.4.4 Trying it out10.5 Adding custom tools10.5.1 The currency converter10.5.2 Trying it out10.6 Indexing multimodal data with LlamaIndex10.6.1 Overview10.6.2 Installing LlamaIndex10.6.3 Implementing a simple question-answering system10.6.4 Trying it out10.7 Concluding remarks10.9 References

Content preview from Data Analysis with LLMs

9 Optimizing cost and quality

This chapter covers

Model choice and tuning
Prompt engineering
Fine-tuning models

Analyzing data with large language models is a great way to burn money quickly. If you’ve been using GPT-4 (or a similarly large model) for a while, you’ve probably noticed how fees pile up quickly, forcing you to recharge your account regularly. But do we always need to use the largest (and most expensive) model? Can’t we make smaller models perform almost as well? How can we get the most bang for our buck?

This chapter is about saving money when using language models on large data sets. Fortunately, we have quite a few options for doing so. First, we have lots of choices when it comes to large language models. Selecting ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Practical Statistics for Data Scientists, 2nd Edition

Publisher Resources

ISBN: 9781633437647Publisher Support Publisher Website

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

Data Analysis with LLMs

by Immanuel Trummer