Data Science And Mlops With Feast: Mastering Feature Store

Unlock the Potential of Feature Store with Feast: Theory and Practice. Configure Feast locally and with Cloud components

What you'll learn

Understanding the concept of Feature Store

Understanding the role of Feast in the Feature Store ecosystem.

Implementing Feast in practice

Integrating Feast with other components

Requirements

Participants should have basic knowledge of machine learning, some familiarity with data technologies, and a foundational understanding of Python. The course is beginner-friendly, welcoming those without prior experience to join and learn from scratch.

Description

Why you shoud learn about Feature Store? Well, it's like having a super tool for organizing important things in your computer. Imagine you have tons of puzzle pieces (data), and you want to build a cool picture (machine learning model). Feature Store helps you keep these puzzle pieces neat and organized, saving a lot of time! Learning about Feature Store is like getting the secret key to make your computer puzzles super easy and fast.Why Feast? Feast stands out as the go-to open-source library in the Feature Store landscape. Its popularity is not without reason—it seamlessly integrates with major cloud platforms like GCP, AWS, and Azure, providing a versatile solution for diverse environments.In this course, discover the 'why' behind Feature Stores and Feast through a structured curriculum:Foundations of Feature Stores: Delve into the core theories underpinning Feature Stores, understanding their pivotal role in modern data management.Feast Essentials: Unpack the unique features of Feast, exploring its functionalities that make it a preferred choice in the open-source community.Cloud Integration Techniques: Learn the art of configuring Feast in cloud environments, gaining practical insights into integration with GCP.Hands-On Practice: Apply your knowledge through hands-on exercises, ensuring you not only comprehend the theories but can adeptly implement them in real-world scenarios.Optimizing Workflows: Discover strategies to optimize data workflows using Feast, enhancing efficiency and performance in machine learning model development.By the end of this course, you'll not only understand the 'whys' but also possess the skills to navigate and thrive in the Feature Store landscape, particularly mastering the art of Feast integration with other components like GCP, Github actions.

Overview

Section 1: Introduction

Lecture 1 Welcome & what you will learn from this course?

Section 2: Theory

Lecture 2 The challenges that arise during the creation of ML models

Lecture 3 How Feature Store can help to solve these problems?

Lecture 4 Feature Store & key components

Section 3: Practice - local enivorment

Lecture 5 Set up environment & first overview of Feast

Lecture 6 Feast objects

Lecture 7 Working with Feast in notebook

Section 4: Practice - cloud enviroment

Lecture 8 Instalation extra packages

Lecture 9 How we can use BigQuery with Feast?

Lecture 10 Redis instance in GCP as Online Store

Lecture 11 Setting CloudSQL instance as remote registry

Section 5: Practice - remote repository

Lecture 12 Feast remote repository & CI/CD pipeline

Section 6: Finally overview on Feast workflow

Lecture 13 Feast workflow local vs central mode

The course is intended for individuals with basic knowledge of machine learning and data technologies, while also being beginner-friendly for those starting from scratch.