The Azure Machine Learning SDK allows data scientists and machine learning engineers to harness the cloud's power in their day-to-day operations. Data science enthusiasts can train their models against scalable compute and GPU clusters on-demand in the cloud, all from the same Jupyter notebook experience.

With the Azure Machine Learning SDK, comes Azure ML pipelines. Machine learning engineers can create a CI/CD approach to their data science tasks by splitting their workflows into pipeline steps. This allows for greater scalability when dealing with large scale data. The separation of functions greatly benefits complex model orchestration as engineers and scientists can focus on one segment at a time. It also provides cost savings by enabling different levels of compute for each step.

In this lab, you will dive into Azure Notebooks and launch a Jupyter notebook to build an Azure ML pipeline that ingests data, trains a model, and deploys a web service.

Learning Objectives

Upon completion of this lab you will be able to:

• Manage a workspace using the Azure Machine Learning SDK
• Pass data between pipelines
• Create and run Azure Notebooks
• Build Azure ML Pipelines for your Data Science workflows

Intended Audience

This lab is intended for:

• Individuals studying to take the Azure DP-100 exam
• Anyone interested in learning how to use the Azure Machine Learning SDK

Lab Prerequisites

You should be familiar with:

• Basic concepts of Azure Machine Learning
• Experience with Python is not required but preferred

Updates

September 8th, 2022 - Update notebook to use the latest azureml notebook environment

August 26th, 2021 - Update notebook code to not use deprecated classes

August 25h, 2021 - Updated instructions to reflect the latest Azure Machine Learning experience