Azure Machine Learning

Question 1

What is Azure Machine Learning?

Answer 1

A cloud-based platform for building and operating ML solutions in Azure.

Question 2

What is Azure ML Studio?

Answer 2

A web portal in Azure ML for low-code and no-code options for model training, deployment, and asset management.

Question 3

What is an Azure ML Workspace?

Answer 3

The top-level resource for Azure ML, providing a centralised place to work with all the artifacts you create when you use Azure ML.

Question 4

What are compute targets?

Answer 4

Cloud-based resources on which you can run model training and data exploration processes.

Question 5

What are the 4 kinds of compute targets?

Answer 5

Compute Instances

Compute Clusters

Inference Clusters

Attached Compute

Question 6

What are Compute Instances?

Answer 6

Development workstations that data scientists can use to work with data and models.

Question 7

What are Compute Clusters?

Answer 7

Scalable clusters of virtual machines for on-demand processing of experiment code.

Question 8

What are Inference Clusters used for?

Answer 8

Deployment targets for predictive services that use your trained models.

Question 9

What are Attached Compute used for?

Answer 9

Links to existing Azure compute resources, such as VMs or Azure Databrick clusters.

Question 10

What is a dataset in Azure ML?

Answer 10

An object that contains data for model training and another operations.

It represents specific data files or tables that you plan to work with in Azure ML.

Question 11

What is automated machine learning in Azure ML?

Answer 11

A feature that leverages the scalability of cloud compute to automatically try multiple pre-processing techniques and model-training algorithms in parallel to find the best performing model.

It only(?) supports supervised ML.

Question 12

What is Azure Cognitive services?

Answer 12

A comprehensive family of AI services and cognitive APIs to help you build intelligent apps.

Question 13

What decision AI services does Azure Cognitive services encapsulate?

Answer 13

Anomaly Detector

Content moderator

Personaliser

Question 14

What language AI services does Azure Cognitive services encapsulate?

Answer 14

Language understanding (LUIS)

QnA Maker

Text analysis

Translator

Question 15

What speech AI services does Azure Cognitive services encapsulate?

Answer 15

Speech-to-Text

Text-to-Speech

Speech translation

Question 16

What vision AI services does Azure Cognitive services encapsulate?

Answer 16

Computer Vision

Custom Vision

Face API

Question 17

For a majority of the services under the Cognitive services, you can create a separate resource to use. Why would you want to use the Cognitive services?

Answer 17

If developers used more than one cognitive service (e.g. Face and LUIS) and wanted to have one key and one endpoint to connect to.

Question 18

How is the Language Understanding service under cognitive services resource different from the stand-alone resource?

Answer 18

In cognitive service, you can only create a prediction Language Understanding resource; if you want authoring, you’ll have to make a stand-alone.

Question 19

What are the 3 settings for creating a compute instance in Azure ML Studio?

Answer 19

Virtual Machine Size
Virtual Machine Type
Compute Name

Question 20

What are the 5 settings for creating a compute cluster in Azure ML Studio?

Answer 20

Virtual Machine Size
Virtual Machine Type
Virtual Machine Priority

Compute Name

Minimum number of nodes
Maximum number of nodes

Idle seconds before scale down

Question 21

What are the 4 settings for deploying a model in Azure ML Studio?

Answer 21

Name
Description
Compute type
Enable authentication

Question 22

What are the 3 resources added automatically to when creating a ML workspace ?

Answer 22

Key vault
Storage account
Application Insights