MLOps is the set of practices developed to help you deploy and maintain machine learning models in production. Nowadays, in industry and research, MLOps is in the spotlight as a way to ensure that ML systems produce value.
Discover Full Automation in MLOpsIn this course, you will learn how to use automation in MLOps to deploy ML systems that can deliver value over time. You'll learn how hidden technical debt affects ML systems and the value they produce. You'll also understand how automating and streamlining the stages of the ML lifecycle can help the operation and scaling of ML systems.
Learn About MLOps ArchitectureYou will use hands-on and interactive exercises to learn about the components of an MLOps architecture and how these are necessary to enable the full automation of ML systems.
Explore CI/CD/CM/CT MLOps TechniquesAs you progress, you’ll learn how automated CI/CD, together with Continuous Monitoring (CM) and Continuous Training (CT), are key techniques to avoid technical debt in your ML deployments.
Understand Automation in Deployment StrategiesBy the end of the course, you’ll understand how automation with MLOps can improve how you deploy your ML systems to the real world, providing your deployments with robustness and scalability.
Start learning, gain knowledge in this highly in-demand field, and discover how to apply automation when designing MLOps systems.
Introduction: to Fully Automated MLOpsFree
In this first chapter, we motivate the use of MLOps in an industrial setting. You’ll learn about its importance in supporting the generation of value in businesses. You’ll also recap the ML stages, focusing on how MLOps enhances these. At the end of the chapter, you’ll explore a reference architecture for a fully automated MLOps system. You will then use this architecture to explore components important for any MLOps system and a starting point for the rest of the course.Introduction to fully automated MLOps50 xpHidden technical debt must be paid50 xpHidden tech debt in ML is different100 xpMLOps lifecycle stages50 xpFully automated MLOps50 xpTo automate or not to automate...50 xpBuilding for scale100 xpReference architecture: Fully automated MLOps50 xpReference architectures in IT and MLOps50 xpMLOps architecture components100 xp
Fully Automated MLOps Architecture
In this chapter, you will continue your exploration of the critical components that make up a fully automated MLOps system. First, you’ll examine the importance of orchestration in MLOps and how it helps to ensure the efficiency and scalability of ML pipelines. After this, you’ll examine the different deployment strategies in MLOps and learn how to choose the right strategy for your system. Finally, you’ll learn about CI/CD/CT/CM and how it complements orchestration and best practices to achieve full automation in MLOps systems. With these lessons under your belt, you will be better equipped to build a fully automated MLOps system that is efficient, accurate, and reliable.Automated experiment tracking50 xpWhat should we keep track of?50 xpThe great experiment tracking debate: Which side are you on?50 xpWhat are we tracking here?100 xpThe model registry50 xpDo not throw the model over the fence!50 xpThe ML model lifecycle100 xpThis is a job for the model registry!100 xpThe feature store in an automated MLOps architecture50 xpAn advantage of using a feature store50 xpThe roles of the feature store100 xpThe metadata store50 xpMetadata in MLOps50 xpModel retraining with the metadata store100 xpThe role of the metadata store in monitoring MLOps100 xp
In this chapter, you’ll dive into the exciting world of automation and learn how to design more resilient and efficient ML systems. You'll start by understanding the fundamentals of automation in MLOps systems and then move on to discover the power of design patterns and fail-safe mechanisms. You'll also learn how to implement automated testing in MLOps systems and how to use hyperparameter tuning to optimize your models and workflows. By the end of this chapter, you'll be equipped with the skills and knowledge necessary to build and manage fully automated MLOps systems that are both efficient and reliable.Introducing full automation and best practices to MLOps50 xpAutomation in MLOps workflows50 xpDegrees of automation in MLOps100 xpThe automation, monitoring, incident response pattern50 xpIs this a design pattern?50 xpRetrain, rollback or input?100 xpAutomated testing in MLOps50 xpGarbage in - Garbage out... Apply data tests!50 xpTesting: Traditional vs. MLOps100 xpTesting: Data vs. model vs. pipeline100 xpAutomated hyperparameter tuning50 xpParameter or hyperparameter?50 xpThe role of hyperparameter tuning in MLOps100 xpAutomating hyperparameter tuning100 xp
Automation in MLOps Deployments
In this final chapter, you’ll delve into the crucial components of an automated MLOps architecture. From understanding automated experiment tracking and the model registry to exploring the feature store and the role of the metadata store, this chapter is designed to equip you with a comprehensive understanding of the intricacies of a fully automated MLOps system. Whether you're a seasoned ML practitioner or just starting out, this chapter will provide you with the knowledge and skills necessary to design automated MLOps workflows.Orchestration in MLOps50 xpWhy should we modularize?50 xpOrchestration: dev vs. prod100 xpDAGs in MLOps workflows50 xpOrchestration aspects in MLOps100 xpAutomation in MLOps deployment strategies50 xpWhat kind of deployment is this?100 xpIs this the right deployment strategy?50 xpKeeping a zero downtime with a blue/green deployment100 xpCI/CD/CT/CM in Fully Automated MLOps50 xpHow does CI/CD/CT/CM contribute to MLOps?50 xpContinuous monitoring & continuous training100 xpCI/CD/CT/CM in MLOps100 xpSummary50 xp
PrerequisitesMLOps Deployment and Life Cycling
Arturo Opsetmoen AmadorSee More
Senior Consultant - Machine Learning
I am an executive MBA with a background as a Ph.D. in physics with research experience. I've worked in the field of data and analytics for several years. I believe that such a field is ever-evolving and broad. This is the reason for my constant-learning obsession. Working as a machine learning engineer with a business degree, I have seen the importance of adopting MLOps best practices to support data strategies. By implementing DevOps in my analytics work, I can deliver data solutions without technical debt. This contributes to sustainable value generation. I like to combine my business knowledge with my technical capabilities. I love to help create data products with a positive business impact.