Deploying and Managing Models in Microsoft Azure Flashcards

1
Q

What is the top level resource for Azure Machine Learning?

Plan

Project

Workspace

Solution

A

Workspace

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are training runs you use to build your models?

Solutions

Sessions

Thoughts

Experiments

A

Experiments

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the last step in an Azure Machine Learning cycle?

Repeat

Deploy

Monitor

Retrain

A

Repeat

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is the term for the change in model input data that leads to model performance degradation?

Function collapse

Data drift

Gradient merge

Model devolution

A

Data drift

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What tool is used to track data drift?

Azure Functions

Azure Application Insights

Azure Container Instances

Azure Security Center

A

Azure Application Insights

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Which tool can be used to integrate Azure Machine Learning with CI/CD?

Azure DevOps

Azure Container Instances

Azure App Service

Azure IoT Edge

A

Azure DevOps

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What can trigger an Azure Machine Learning automated deployment?

Tensorflow Checkpoint

Git Checkin

Azure Function

Database Update

A

Git Checkin

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is the type of message used to monitor for data drift?

data_drift

model_data_collection

drift_detection

accuracy_monitor

A

model_data_collection

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the statistical distance defined for one-dimensional numerical distribution?

energy_distance

wasserstein_distance

datadrift_contribution

datadrift_coefficient

A

wasserstein_distance / energy_distance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is the statistical distance defined for one-dimensional numerical distribution?

datadrift_coefficient

datadrift_contribution

energy_distance

wasserstein_distance

A

wasserstein_distance / energy_distance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly