Building Machine Learning Systems with a Feature Store (Early Release)

✍ Scribed by Jim Dowling

• Publisher: O'Reilly Media, Inc.
• Year: 2024
• Tongue: English
• Leaves: 39
• Category: Library

✦ Synopsis

Get up to speed on a new unified approach to building machine learning (ML) systems with batch data, real-time data, and large language models (LLMs) based on independent, modular ML pipelines and a shared data layer. With this practical book, data scientists and ML engineers will learn in detail how to develop, maintain, and operate modular ML systems. Author Jim Dowling introduces fundamental MLOps principles and practices for developing and operating reliable ML systems and describes the key data platform that you'll use to build and operate your ML systems: the feature store.

✦ Table of Contents

Brief Table of Contents (Not Yet Final)
Preface
Overview of this book’s mission
Target Reader of this Book
Introduction
From ML Models to MLOps to ML Systems
Supervised learning primer and what is a feature anyway?
1. Building Machine Learning Systems
The Evolution of Machine Learning Systems
The Anatomy of a Machine Learning System
Types of Machine Learning
Data Sources
Tabular data
Unstructured Data
Event Data
API-Provided Data
Ethics and Laws for Data Sources
Incremental Datasets
What is a ML Pipeline ?
Principles of MLOps
Machine Learning Systems with a Feature Store
Three Types of ML System with a Feature Store
ML Frameworks and ML Infrastructure used in this book
Summary