For readers of the O’Reilly book
3D Data Science with Python
Everything the book points to, in one place: the code and data for all 18 chapters, 31 Resource Hub guides and a 3D preview of every dataset. Enter your email and the link lands in your inbox.
Written by Dr. Florent Poux.
O’Reilly, 2025. 690 pages. 15,000+ readers.
ISPRS Jack Dangermond Award, 1,700+ citations.
Get the book companion resources
The link to the code and data for all 18 chapters, plus a short series on putting the book to work.
What you get
Built around the book, chapter by chapter
18 chapter pages
What each chapter builds, its files, and the exact command to run it.
Code and data
One folder per chapter, code and datasets side by side, fixed for current library versions.
The Resource Hub
The 31 guides the book sends you to, from sensors to SDF meshing to metrics.
3D previews
Orbit every chapter’s dataset in your browser before you download a single gigabyte.
Every chapter covered
One page per chapter, waiting for you
Each one lists what you build, the files, the run command and its Resource Hub guides. The link to all 18 comes with your email.
Questions readers ask
Before you start
Where are the code and the datasets?
In one shared folder with a CHAPTER_NN subfolder per chapter, each holding CODE and DATA side by side. Enter your email above: the link to that folder and to the companion, where every chapter page lists its files, arrives in your inbox.
The code fails on my machine. What now?
Create the DATAS3D environment from the environment.yml file in the shared folder: versions are pinned and every chapter was re-run against them. The companion's Setup page and each chapter page note any file names that differ from the print.
Which Python version, and do I need a GPU?
Python 3.10. Chapters 1 to 12, 14 and 17 run on any laptop. Segment Anything (chapter 13), graph deep learning (chapter 15) and 3D generative AI (chapter 18) are faster with a CUDA GPU and are optional.
How big are the datasets?
The code is small. Datasets range from a few megabytes to about 500 MB per file, so download only the chapters you are working on. Each chapter page shows file sizes and a 3D preview first.
Keep going
Where readers go next
Free
The blog
Practical tutorials and deep dives on 3D and spatial AI you can drop into your projects.
Free, daily
3D Intelligence Reports
A paper, a tool, a job and a funding move in 3D spatial AI, verified the same day.
Self-paced
The course Library
Focused micro-courses, blueprints and complete operating systems for 3D AI. One subscription, or one course at a time.
Flagship program
The 3D AI Architect Program
The guided path from book reader to 3D AI architect who designs and ships production systems.
Go further than the book
Build production 3D AI with Florent
The 3D AI Architect Program is the guided path from book reader to 3D AI architect who designs and ships production systems.
