What do you think?
Rate this book
Building Machine Learning Powered Applications: Going from Idea to Product
Learn the skills necessary to design, build, and deploy applications powered by machine learning. Through the course of this hands-on book, you'll build an example ML-driven application from initial idea to deployed product. Data scientists, software engineers, and product managers with little or no ML experience will learn the tools, best practices, and challenges involved in building a real-world ML application step-by-step.
Author Emmanuel Ameisen, who worked as a data scientist at Zipcar and led Insight Data Science's AI program, demonstrates key ML concepts with code snippets, illustrations, and screenshots from the book's example application.
The first part of this guide shows you how to plan and measure success for an ML application. Part II shows you how to build a working ML model, and Part III explains how to improve the model until it fulfills your original vision. Part IV covers deployment and monitoring strategies.
This book will help you:
Determine your product goal and set up a machine learning problem
Build your first end-to-end pipeline quickly and acquire an initial dataset
Train and evaluate your ML model and address performance bottlenecks
Deploy and monitor models in a production environment
Author Emmanuel Ameisen, who worked as a data scientist at Zipcar and led Insight Data Science's AI program, demonstrates key ML concepts with code snippets, illustrations, and screenshots from the book's example application.
The first part of this guide shows you how to plan and measure success for an ML application. Part II shows you how to build a working ML model, and Part III explains how to improve the model until it fulfills your original vision. Part IV covers deployment and monitoring strategies.
This book will help you:
Determine your product goal and set up a machine learning problem
Build your first end-to-end pipeline quickly and acquire an initial dataset
Train and evaluate your ML model and address performance bottlenecks
Deploy and monitor models in a production environment
256 pages, Paperback
Published February 11, 2020
108 people are currently reading
616 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 14 of 14 reviews
February 24, 2020
This book is not about the fad of machine learning but what else comes with it.
In current times, ML has become something which most of us think that we need to create some model, learn some algorithm and it will be done, which is not true in the real world.
This book explains what are the different things that come with a simple problem which everyone thinks can be solved with ML algorithms. I think everyone who wants to work on machine learning projects should read this book. It's a good and quick read and can be referred back to again and again.
Thanks Emmanuel.
In current times, ML has become something which most of us think that we need to create some model, learn some algorithm and it will be done, which is not true in the real world.
This book explains what are the different things that come with a simple problem which everyone thinks can be solved with ML algorithms. I think everyone who wants to work on machine learning projects should read this book. It's a good and quick read and can be referred back to again and again.
Thanks Emmanuel.
June 1, 2020
Building Machine Learning Powered Applications: Going from Idea to Product
By Emmanuel Ameisen
This is not the typical machine learning book that I have seen so far that covers the various learning algorithms and models. It takes the reader from a product idea, with a practical example, in a walkthrough style of writing, all the way to the deployment. It provides the full source code of the application built and it is discussed in the book chapters. As it states in the book, it is not an introductory book to machine learning but requires some basic knowledge in the area. Reading the book - Building Machine Learning powered application, going from idea to product - feels like working in a real world project with a mentor giving advice on little details that could go wrong and how to be fix it.
In addition to the practical examples the author provides in the discussion of the individual topics, the interviews in the book also reinforce the concepts using experiences of various professionals from multiple big organisations. The references cited in the book also supplement the material discussed. For me, this book is what has been missing to transition from playing with existing datasets to experiment with machine learning models in class or other books to the practical world where you validate the idea, collect and prepare the data, build a complete pipeline, evaluate your model, deploy and monitor the output of your model to refine it in production and iterate.
In general, the book is perfect reference for individuals looking a carrier in companies building ML powered applications or companies that are planning to integrate ML in their product offering. I am sure I have to get back and refer to it when I am going to work on my project soon.
March 30, 2020
This book answered so many questions I had about a transition between an ML playground experiment to having an ML-powered product. Lots of practical examples mixed with insightful interviews. Extremely glad I picked this book up!
December 27, 2021
Reading this book as a novice in the deployment side of datascience I have mixed feelings about the book. I especially enjoyed and learned a lot from the first chapters on ideation/design/data requirements and exploration and finally model training. In these chapters I came across perspectives and tips on these processes I havent seen as clearly explained in other books, and there are multiple notes that I will use in future projects here. Therefore I had high expectations for the latter chapters on deployment and management of ML applications, which were unfortunately to generalistic to my taste. I think it is really fine to do a birds eye view of these aspects as long as you provide plenty of examples of deployment options. I also missed some more information on different deployment areas including cloud providers. In my opinion next iterations of this book could benefit from: A more elaborate GitHub with multiple applications, more practical and in depth advice on deployment chapters and possibly a more commonly used application then the NLP example to start with. All in all definitely worth the read, therefore 4 stars.
Read this book if: you want a birds eye view of every aspect of the lifecycle of an ML application from planning, data gathering to deployment and maintenance
Do not read this book: If you want to have strong applied examples.
Read this book if: you want a birds eye view of every aspect of the lifecycle of an ML application from planning, data gathering to deployment and maintenance
Do not read this book: If you want to have strong applied examples.
April 18, 2020
💫The book came in a very timely fashion for me. As a 1-man ML team, it's easy to get stuck in a place where you're not sure what the next step is.
It's "easy" to learn (and find books about) the theoretical part of ML modelling, or how to rank higher in Kaggle. But "what is much rarer, however, is the ability to take a problem, estimate how best to solve it, build a plan to tackle it with ML, and confidently execute on said plan. This is often a skill that has to be learned through experience, after multiple overly ambitious projects and missed deadlines".
I myself have been battling with one such project for almost a year now (been doing other stuff too but still). This book helped me go back and understand where I failed prioritising the right "performance bottleneck" 💡. This book also helped me validate what I felt were the challenges for ML powered applications, even when you feel alone in an organisation, having to defend your progress for example, or explain why the outcome of your model is not deterministic (🤦♂️).
It's "easy" to learn (and find books about) the theoretical part of ML modelling, or how to rank higher in Kaggle. But "what is much rarer, however, is the ability to take a problem, estimate how best to solve it, build a plan to tackle it with ML, and confidently execute on said plan. This is often a skill that has to be learned through experience, after multiple overly ambitious projects and missed deadlines".
I myself have been battling with one such project for almost a year now (been doing other stuff too but still). This book helped me go back and understand where I failed prioritising the right "performance bottleneck" 💡. This book also helped me validate what I felt were the challenges for ML powered applications, even when you feel alone in an organisation, having to defend your progress for example, or explain why the outcome of your model is not deterministic (🤦♂️).
December 20, 2020
By it's title and introduction, the book seemed to me as book for software developer who is new to machine learning (which I am). I expected ML algorithms to be described from development view (algorithms and suggestions how and when to use them). It appeared this book almost doesn't contain algorithms. Instead it takes one sample algorithm and shows how to use it from scratch. I don't think this book can be helpful as a first ML book. From my point of view, it also lacks depth needed for software development. Maybe it can be useful for someone but I can hardly imagine who is the reader.
December 9, 2023
I know a little bit about modeling, data science and mlops, but this books showed me a whole world of aspects to take into account to add machine learning to applications, where modeling is a tiny part.
It's very comprehensive and it has a lot of examples
It's not specially deep in any aspect, but I think the intention here is to do a pass through everything and then let the user further zoom in in each spect
It's very comprehensive and it has a lot of examples
It's not specially deep in any aspect, but I think the intention here is to do a pass through everything and then let the user further zoom in in each spect
October 27, 2020
This book is a must-read. Especially for those that are new to the field. It tackles questions most other ML books overlook and exposes the messy details and enormous work that is required to deliver ML products.
Currently reading
March 30, 2021This book contains lots of practical tips to succeed in solving a problem with machine learning. The companion repository for the book is also very helpful to put into practice the main ideas presented.
Totally recommended for beginners and experienced ML practitioners alike!
Totally recommended for beginners and experienced ML practitioners alike!
January 7, 2024
It's a good book covering the ML product life cycle with code included. It's a good reference for learning more about ML and working on ML product projects. The author is nice too, he would answer questions when you reach out to him.
October 14, 2021
A practical overview with examples of building end-to-end training and inference pipelines. Might be a good intro for a person who wants to understand how it's done but is insufficient to help to build. What is unique about this book that it covers how to prepare, test data and debug models in practical details.
June 9, 2022
The book was authored by Emmanuel Ameisen, a Machine Learning Engineer in Stripe. The book is very handy, and it goes on a journey of Machine Learning powered products, starting from the idea to the product. It's beneficial for the people who want to have an ML-powered product on the market; it gives a bird's view of the significant aspects of the whole journey.
The book describes in detail the process of the whole journey, which includes:
* Defining the use-case
* Creating a plan
* Acquiring and collecting Data
* Selecting the proper model
* Building a pipeline
* Deploying, Monitoring, and evaluating model
* Software / ML Best Practices
All in all, it's beneficial for the people in the field, including Product Managers, Machine learning Engineers, Practitioners, and Software Engineers.
The book describes in detail the process of the whole journey, which includes:
* Defining the use-case
* Creating a plan
* Acquiring and collecting Data
* Selecting the proper model
* Building a pipeline
* Deploying, Monitoring, and evaluating model
* Software / ML Best Practices
All in all, it's beneficial for the people in the field, including Product Managers, Machine learning Engineers, Practitioners, and Software Engineers.
Read
January 6, 2025I have a short review here: https://blog.revfad.com/2024/04/29/sh...
Displaying 1 - 14 of 14 reviews