MLOps for Azure Machine Learning

• Experience with Azure Machine Learning

Audience

• Data Scientists

MLOps (Machine Learning Operations) is the practice of integrating data science and operations to help manage the ML lifecycle. MLOps provides the ability to automate the reproduction of machine learning model development and training.

This instructor-led, live training (online or onsite) is aimed at data scientists who wish to use Azure Machine Learning and Azure DevOps to facilitate MLOps practices.

By the end of this training, participants will be able to:

• Build reproducible workflows and machine learning models.
• Manage the machine learning lifecycle.
• Track and report model version history, assets, and more.
• Deploy production ready machine learning models anywhere.

Format of the Course

• Interactive lecture and discussion.
• Lots of exercises and practice.
• Hands-on implementation in a live-lab environment.

Course Customization Options

• To request a customized training for this course, please contact us to arrange.

Introduction

MLOps Overview

• What is MLOps?
• MLOps in Azure Machine Learning architecture

Preparing the MLOps Environment

• Setting up Azure Machine Learning

Model Reproducibility

• Working with Azure Machine Learning pipelines
• Bridging Machine Learning processes with pipelines

Containers and Deployment

• Packaging models into containers
• Deploying containers
• Validating models

Automating Operations

• Automating operations with Azure Machine Learning and GitHub
• Retraining and testing models
• Rolling out new models

Governance and Control

• Creating an audit trail
• Managing and monitoring models

Summary and Conclusion