Thoughts - Ambarish

01 Nov 2020

Azure ML workspace basics

Azure ML has a top level component which is the Workspace. The workspace contains all the components of the Azure Machine Learning space.

The workspace is associated with

  • Azure subscription
  • Azure Key Vault
  • Azure Application Insights

It is associated with the following assets

  • Datasets
  • Experiments
  • Pipelines
  • Models
  • EndPoints

The workspace manages the following

  • Datastores
  • Compute

The workspace can be used for authoring

  • Notebooks
  • Automated ML
  • Designer

You can implement via code

import azureml.core
print(azureml.core.VERSION)
from azureml.core import Workspace
from azureml.core.authentication import InteractiveLoginAuthentication

sid = <your-subscription-id>
forced_interactive_auth = InteractiveLoginAuthentication(tenant_id=<your-tenant-id>)
ws = Workspace.create(name='azureml_workspace',
            subscription_id= sid, 
            resource_group='rgazureml',
            create_resource_group = True,
            location='centralus'
            )