Managing Machine Learning Projects: From design to deployment - Softcover

Thompson, Simon

Softcover

ISBN 10: 163343902X ISBN 13: 9781633439023

Publisher: Manning, 2023

Synopsis

Guide machine learning projects from design to production with the techniques in this unique project management guide. No ML skills required!

In Managing Machine Learning Projects you’ll learn essential machine learning project management techniques, including:

• Understanding an ML project’s requirements
• Setting up the infrastructure for the project and resourcing a team
• Working with clients and other stakeholders
• Dealing with data resources and bringing them into the project for use
• Handling the lifecycle of models in the project
• Managing the application of ML algorithms
• Evaluating the performance of algorithms and models
• Making decisions about which models to adopt for delivery
• Taking models through development and testing
• Integrating models with production systems to create effective applications
• Steps and behaviors for managing the ethical implications of ML technology

Managing Machine Learning Projects is an end-to-end guide for delivering machine learning applications on time and under budget. It lays out tools, approaches, and processes designed to handle the unique challenges of machine learning project management. You’ll follow an in-depth case study through a series of sprints and see how to put each technique into practice. The book’s strong consideration to data privacy, and community impact ensure your projects are ethical, compliant with global legislation, and avoid being exposed to failure from bias and other issues.

Purchase of the print book includes a free eBook in PDF, Kindle, and ePub formats from Manning Publications.

About the Technology

Ferrying machine learning projects to production often feels like navigating uncharted waters. From accounting for large data resources to tracking and evaluating multiple models, machine learning technology has radically different requirements than traditional software. Never fear! This book lays out the unique practices you’ll need to ensure your projects succeed.

About the Book

Managing Machine Learning Projects is an amazing source of battle-tested techniques for effective delivery of real-life machine learning solutions. The book is laid out across a series of sprints that take you from a project proposal all the way to deployment into production. You’ll learn how to plan essential infrastructure, coordinate experimentation, protect sensitive data, and reliably measure model performance. Many ML projects fail to create real value—read this book to make sure your project is a success.

What's Inside

• Set up infrastructure and resource a team
• Bring data resources into a project
• Accurately estimate time and effort
• Evaluate which models to adopt for delivery
• Integrate models into effective applications

About the Reader

For anyone interested in better management of machine learning projects. No technical skills required.

About the Author

Simon Thompson has spent 25 years developing AI systems to create applications for use in telecoms, customer service, manufacturing and capital markets. He led the AI research program at BT Labs in the UK, and is now the Head of Data Science at GFT Technologies.

Table of Contents

1 Introduction: Delivering machine learning projects is hard; let’s do it better
2 Pre-project: From opportunity to requirements
3 Pre-project: From requirements to proposal
4 Getting started
5 Diving into the problem
6 EDA, ethics, and baseline evaluations
7 Making useful models with ML
8 Testing and selection
9 Sprint 3: system building and production
10 Post project (sprint O)

About this title

About the Author

Simon Thompson has spent 30 years developing AI systems. He led the AI research program at BT Labs in the UK, where he helped pioneer Big Data technology in the company and managed an applied research practice for nearly a decade. Simon now works delivering Machine Learning systems for financial services companies in the City of London as the Head of Data Science at GFT Technologies.

From the Back Cover

Ferrying machine learning projects to production often feels like navigating uncharted waters. From accounting for large data resources to tracking and evaluating multiple models, machine learning technology has radically different requirements from traditional software. Never fear! This book lays out the unique practices you'll need to ensure your projects succeed.

From the Inside Flap

I can't pin down a moment, or weave a convincing anecdote that explains how I came to realize that writing a book about how to manage a machine learning project would be a good thing to do. The gist of it is that sometime in 2019 I realized that I was talking to a lot of people who had started an ML project and were in trouble with it, and usually I knew why.

There wasn't one common malady or even a single theme, rather failures seemed to come from lots of different directions. Disparate as the failings of these projects were there was a common cause at work here. The folks leading these projects were talented, clever, articulate and skilled, but they were inexperienced.

I was very lucky in the timing of my career. I got into ML when it was on the edge of applications. In the late 1990's ML was out there in the wild, and we could do real things with our three-layer perceptron's and decision trees. It was much harder to deliver though, algorithms needed to be coded by hand, data was vanishing rare and everything ran sooooo slowly. Most of all, ML skills were as rare as the projects that needed them. Most of all, applied ML was seen as R&D. For me this meant that I had the opportunity to develop and work on project after project. Most of them failed - but the ones that did come off really, really, really came off.

At the time the rare wins kept me in work, and kept my career going. In turn this paid the mortgage and filled the freezer. With hindsight I can say now that it was the failures that were the most valuable. I had the luxury of failure and learning, which isn't often afforded to people today. I also got the opportunity to join communities of people going through the same thing, and we would all get really drunk and tell each other sad (and funny) stories of catastrophe. A bunch of practices and behaviours became common knowledge in the clique of AI researchers working in big western companies in those days. I had the luck of being able to pick this all up and then use it.

Having had the luck of getting enough experience to steer an ML project or two to success it would be churlish not to share it. I firmly believe that ML and AI are technologies that can be used for good, hopefully helping people confronted with climate change, pandemics and economic woes. Maybe by sharing knowledge about how to manage ML projects I can help someone else do a couple of projects that make the world a better place!

Two events pushed the book from an idea into the real world though. First Andy Rossiter who was my boss at the time told me we needed to have a methodology to tell customers how we would tackle their problems. I realized that I couldn't really point at one so I'd have to write one. That probably wouldn't have gone all that far if it wasn't for the second event - the pandemic - that meant that I stopped spending hours travelling about and started to have some time in the evening to commit to writing something.

So, here it is. Thank you for buying it. I hope you find it useful and most of all I hope you will share any ideas or thoughts you have for how it should be improved so that we can do better next time.

Edition details

• Publisher: Manning
• Publication date: 2023
• Language: English
• ISBN 10: 163343902X
• ISBN 13: 9781633439023
• Binding: Paperback
• Number of pages: 272