Creating Machine Learning Applications: From Idea to Product

Creating Machine Learning Applications: From Idea to Product

Master the key skills of designing, developing, and deploying machine learning (ML) applications! A step-by-step guide to building ML applications with a focus on practice: for data processing specialists, software developers, and product managers. By reading this book, you will... create a real practical application step by step — from idea to implementation. You will have access to code samples, illustrations, screenshots, and interviews with leading experts in the field. You will learn to plan and measure the success of ML projects, understand how to build a working model, and master ways to iteratively refine it. Finally, you will become familiar with deployment and monitoring strategies.

The book will help you:
• Define the goal of your ML project
• Quickly build the first end-to-end pipeline and find the initial dataset
• Train the model and eliminate performance bottlenecks
• Deploy the model and monitor it

Author: Эммануэль Амейзен

Printhouse: piter

Series: Programmer's Library

Age restrictions: 16+

Year of publication: 2023

ISBN: 9785446117734

Number of pages: 272

Size: 233x165x15 mm

Cover type: soft

Weight: 440 g

ID: 1508997

Master the key skills of designing, developing, and deploying machine learning (ML) applications! A step-by-step guide to building ML applications with a focus on practice: for data processing specialists, software developers, and product managers. By reading this book, you will create a real practical application step by step — from idea to implementation. You will have access to code samples, illustrations, screenshots, and interviews with leading experts in the field. You will learn to plan and measure the success of ML projects, understand how to build a working model, and master ways to iteratively refine it. Finally, you will become familiar with deployment and monitoring strategies.

The book will help you:
• Define the goal of your ML project
• Quickly build the first end-to-end pipeline and find the initial dataset
• Train the model and eliminate performance bottlenecks
• Deploy the model and monitor it