Microsoft Role-based AI-100 Questions & Answers

• Question 1:

  Note: This question is part of a series of questions that present the same scenario. Each question in the series contains a unique solution that might meet the stated goals. Some question sets might have more than one correct solution, while

  others might not have a correct solution.

  After you answer a question, you will NOT be able to return to it. As a result, these questions will not appear in the review screen.

  You have Azure IoT Edge devices that generate streaming data.

  On the devices, you need to detect anomalies in the data by using Azure Machine Learning models. Once an anomaly is detected, the devices must add information about the anomaly to the Azure IoT Hub stream.

  Solution: You expose a Machine Learning model as an Azure web service.

  Does this meet the goal?

  A. Yes

  B. No

  Correct Answer: B

  Instead use Azure Stream Analytics and REST API.

  Note. Available in both the cloud and Azure IoT Edge, Azure Stream Analytics offers built-in machine learning based anomaly detection capabilities that can be used to monitor the two most commonly occurring anomalies: temporary and

  persistent.

  Stream Analytics supports user-defined functions, via REST API, that call out to Azure Machine Learning endpoints.

  References:

  https://docs.microsoft.com/en-us/azure/stream-analytics/stream-analytics-machine-learning-anomaly-detection

• Question 2:

  Note: This question is part of a series of questions that present the same scenario. Each question in the series contains a unique solution that might meet the stated goals. Some question sets might have more than one correct solution, while

  others might not have a correct solution.

  After you answer a question, you will NOT be able to return to it. As a result, these questions will not appear in the review screen.

  You create several AI models in Azure Machine Learning Studio.

  You deploy the models to a production environment.

  You need to monitor the compute performance of the models.

  Solution: You enable Model data collection.

  Does this meet the goal?

  A. Yes

  B. No

  Correct Answer: A

  You need to enable Model data collection.

  References: https://docs.microsoft.com/en-us/azure/machine-learning/service/how-to-enable-data-collection

• Question 3:

  Note: This question is part of a series of questions that present the same scenario. Each question in the series contains a unique solution that might meet the stated goals. Some question sets might have more than one correct solution, while

  others might not have a correct solution.

  After you answer a question, you will NOT be able to return to it. As a result, these questions will not appear in the review screen.

  You create several AI models in Azure Machine Learning Studio.

  You deploy the models to a production environment.

  You need to monitor the compute performance of the models.

  Solution: You enable AppInsights diagnostics.

  Does this meet the goal?

  A. Yes

  B. No

  Correct Answer: B

  You need to enable Model data collection.

  References: https://docs.microsoft.com/en-us/azure/machine-learning/service/how-to-enable-data-collection

• Question 4:

  Note: This question is part of a series of questions that present the same scenario. Each question in the series contains a unique solution that might meet the stated goals. Some question sets might have more than one correct solution, while

  others might not have a correct solution.

  After you answer a question, you will NOT be able to return to it. As a result, these questions will not appear in the review screen.

  You create several AI models in Azure Machine Learning Studio.

  You deploy the models to a production environment.

  You need to monitor the compute performance of the models.

  Solution: You write a custom scoring script.

  Does this meet the goal?

  A. Yes

  B. No

  Correct Answer: B

  You need to enable Model data collection.

  References: https://docs.microsoft.com/en-us/azure/machine-learning/service/how-to-enable-data-collection

• Question 5:

  Your company has recently purchased and deployed 25,000 IoT devices.

  You need to recommend a data analysis solution for the devices that meets the following requirements:

  1.

  Each device must use its own credentials for identity.

  2.

  Each device must be able to route data to multiple endpoints.

  3.

  The solution must require the minimum amount of customized code. What should you include in the recommendation?

  A. Microsoft Azure Notification Hubs

  B. Microsoft Azure Event Hubs

  C. Microsoft Azure IoT Hub

  D. Microsoft Azure Service Bus

  Correct Answer: C

  An IoT hub has a default built-in endpoint. You can create custom endpoints to route messages to by linking other services in your subscription to the hub. Individual devices connect using credentials stored in the IoT hub's identity registry.

  References: https://docs.microsoft.com/en-us/azure/iot-hub/iot-hub-devguide-security

• Question 6:

  You create an Azure Machine Learning Studio experiment.

  You plan to publish the experiment as a Machine Learning Web service.

  You need to ensure that you can consume the web service from Microsoft Excel spreadsheets.

  What should you use?

  A. a Batch Execution Service (BES) and an API key

  B. a Batch Execution Service (BES) and an Azure managed identity

  C. a Request-Response Service (RRS) and an Azure managed identity

  D. a Request-Response Service (RRS) and an API key

  Correct Answer: D

  Steps to Add a New web service

  1.

  Deploy a web service or use an existing Web service.

  2.

  Click Consume.

  3.

  Look for the Basic consumption info section. Copy and save the Primary Key and the Request-Response URL.

  4.

  In Excel, go to the Web Services section (if you are in the Predict section, click the back arrow to go to the list of web services).

  5.

  Click Add Web Service.

  6.

  Paste the URL into the Excel add-in text box labeled URL.

  7.

  Paste the API/Primary key into the text box labeled API key.

  8.

  Click Add.

  References: https://docs.microsoft.com/en-us/azure/machine-learning/studio/excel-add-in-for-web-services

• Question 7:

  You are building an Azure Analysis Services cube for your AI deployment.

  The source data for the cube is located in an on premises network in a Microsoft SQL Server database.

  You need to ensure that the Azure Analysis Services service can access the source data.

  What should you deploy to your Azure subscription?

  A. a site-to-site VPN

  B. a data gateway

  C. Azure Data Factory

  D. a network gateway

  Correct Answer: B

  From April 2017 onward we can use On-premises Data Gateway for Azure Analysis Services. This means you can connect your Tabular Models hosted in Azure Analysis Services to your on-premises data sources through On-premises Data Gateway.

  

  References: https://biinsight.com/on-premises-data-gateway-for-azure-analysis-services/

• Question 8:

  A data scientist deploys a deep learning model on an Fsv2 virtual machine.

  Data analysis is slow.

  You need to recommend which virtual machine series the data scientist must use to ensure that data analysis occurs as quickly as possible.

  Which series should you recommend?

  A. ND

  B. B

  C. DC

  D. Ev3

  Correct Answer: A

  The N-series is a family of Azure Virtual Machines with GPU capabilities. GPUs are ideal for compute and graphics-intensive workloads, helping customers to fuel innovation through scenarios like high-end remote visualisation, deep learning and predictive analytics.

  The ND-series is focused on training and inference scenarios for deep learning. It uses the NVIDIA Tesla P40 GPUs. The latest version - NDv2 - features the NVIDIA Tesla V100 GPUs.

  References: https://azure.microsoft.com/en-in/pricing/details/virtual-machines/series/

• Question 9:

  You have Azure IoT Edge devices that generate measurement data from temperature sensors. The data changes very slowly.

  You need to analyze the data in a temporal two-minute window. If the temperature rises five degrees above a limit, an alert must be raised. The solution must minimize the development of custom code.

  What should you use?

  A. A Machine Learning model as a web service

  B. an Azure Machine Learning model as an IoT Edge module

  C. Azure Stream Analytics as an IoT Edge module

  D. Azure Functions as an IoT Edge module

  Correct Answer: C

  References: https://docs.microsoft.com/en-us/azure/iot-edge/tutorial-deploy-stream-analytics

• Question 10:

  Note: This question is part of a series of questions that present the same scenario. Each question in the series contains a unique solution that might meet the stated goals. Some question sets might have more than one correct solution, while

  others might not have a correct solution.

  After you answer a question, you will NOT be able to return to it. As a result, these questions will not appear in the review screen.

  You are deploying an Azure Machine Learning model to an Azure Kubernetes Service (AKS) container.

  You need to monitor the scoring accuracy of each run of the model.

  Solution: You modify the scoring file.

  Does this meet the goal?

  A. Yes

  B. No

  Correct Answer: B

  Modifying the scoring file does not meet the goal of monitoring the scoring accuracy of each run of the model. Instead, you can use Azure Machine Learning's monitoring capabilities to monitor the scoring accuracy of each run of the model. You can use the Azure Machine Learning SDK to enable monitoring for your models and configure the metrics to monitor. You can also use Azure Machine Learning's Application Insights integration to monitor the performance and usage of your models. You can use the Azure Machine Learning SDK to log custom metrics and events from your scoring script, and you can use the Azure Machine Learning Studio to view and analyze the logs. You can also use Azure Machine Learning's model explainability features to understand how your model is making predictions and identify potential issues with the model