Where do you start when migrating your machine learning workflows to Azure? How do you get the necessary resources provisioned? What resources will you even need? Azure Machine Learning workspaces help answer all of these questions. This lab will have you create an Azure Machine Learning workspace, which is the first step in doing machine learning on Azure.
## Scenario
As your team migrates to the cloud, you need to provision Azure resources to allow you to train machine learning models. You’ll need Storage and Compute, a way to monitor your progress, and a place to work on your Notebooks. Azure provides all of these in one place: Machine Learning workspaces.
You will practice setting up a Machine Learning Studio Workspace for future use.
Learning Objectives
Successfully complete this lab by achieving the following learning objectives:
- Provision the Machine Learning Workspace
- Create a Machine Learning Studio workspace. Look at the Infrastructure-as-Code template for the workspace so you can begin to automate this process.
- View the Resource Group to see all resources created.
- Create a Notebook to Run Code
- Create a new Compute Instance to host your notebook. The Virtual Machine will only be used for the notebook, so use the lowest power machine possible to save you money.
- Create a new Python notebook using the Python AzureML kernel.
- Print "Hello, world!" in the notebook to verify that it is working.
- Stop the Notebook Server
- Stop the Compute Instance being used to run the Jupyter notebook.
