MLOps on Azure training

Learn to apply MLOps on Azure

In this training, we cover the end-to-end process of going from notebook to production. You start by writing an ML pipeline in Azure ML, and learn how to track experiments. You will also learn how to build, containerize, and deploy an API to serve your model to the end user. After that, you will add monitoring to inspect your model performance and API health. Finally, the focus will be on enabling automated deployment with CI/CD.

This training course is aimed at people with Python skills and general ML experience who want to take their solutions to the next level.

What will you learn in this MLOps on Azure training?

After the training, you'll be able to:

• Understand all the necessary components in an end-to-end ML system.
• Create machine learning pipelines in Azure ML.
• Deploy your model as scalable API as Azure ML Endpoint or Azure Container Instance.
• Set up monitoring for your application using Application Insights.
• Integrate and deploy all code through a CI/CD pipeline with GitHub Actions.

Key takeaways

• Key principles of MLOps.
• How to design solutions for deploying models to the cloud.
• How to build an ML pipeline for reproducibly training a model.
• How to automatically deploy and schedule ML pipelines for automated re-training.
• How to track models and metrics in a model library.
• How to deploy and monitor an ML model as scalable containerized REST API.
• How to automate all components through CI/CD.
• Practical advice on how productionize ML models in the cloud.

Program

Day 1

• Key MLOps principles.
• Creating a solution design.
• Getting started with the cloud tooling.
• Training jobs and pipelines.

Day 2

• Experiment tracking.
• Deploying and triggering pipelines.
• Exposing model as containerized API.
• Deploy API to the cloud.

Day 3

• Monitoring training pipelines and online model performance.
• Extend ML pipeline with automated deployment.
• Continuous integration and deployment (CICD) workflows.

Who is it for?

This training is for you if you:

• Want to incorporate best practices from Software Engineering.
• Want to learn more about the cloud and MLOps.

Requirements

You can follow this class if you:

• Already have a solid understanding of Machine Learning, and want to take your models further than the development phase.
• Already have basic SWE skills (basic understanding of Python, Git, Docker).
• Remember, this course is oriented toward ML engineering!

Learning journey & recommended learning

If this training doesn't fit your demands, you can check the following related courses.

• If you'd like to learn more about developing ML models, check - for instance - the Certified Data Science with Python or the Advanced Data Science with Python training courses.
• For more basic programming experience, you can start with our Python for Data Analysis course.

Why should I follow the MLOps on Azure training?

This training will walk you through how to deploy machine-learning applications with one of the public cloud providers and how to automate processes with best engineering practices along the way. You will receive conceptual explanations and practical exercises from industry experts in a highly interactive environment.

What else should I know?

• All literature and course materials are included in the price.
• After registering for this course, you will receive a confirmation email with practical information.