Create Azure Machine Learning Workspace

As a first step, we need to create a Azure Machine Learning (AML) workspace. It contains many resources we will reference later and act as a registration points of computes/resources we will register and utilize.

AML workspace should be created as any other Azure resource. Please navigate to portal.azure.com, make sure you are in the right workspace/directory and click on the plus.

Fig. 1 - Create a resource

Select "Azure Machine Learning"

Fig. 2 - Select AML

Click create on the next page and now you should be met with page such as:

Fig. 3 - AML workspace creation form

Here you need to pick the right subscription and region. Your billing and latencies can depend on the region, so make sure to pick the right one.

You can generate other things and their names, unless you are instructed otherwise by the project people. In our opinion, it is easiest to just pick a name of the workspace and do something like workspace-name + - + acronym of the thing, resulting in resource group being workspace-name-rg or generated by Azure.

Afterwards the AML workspace is created, you can move forward.