You manage an Azure Machine Learning workspace named Workspace1.
You plan to create a pipeline in the Azure Machine Learning Studio designer. The pipeline must include a custom component.
You need to ensure the custom component can be used in the pipeline.
What should you do first?
A. Create a pipeline endpoint.
B. Add a linked service to Workspace1.
C. Upload a .json file to Workspace1.
D. Create a datastore.
E. Upload a .yaml file to Workspace1.
Show Answer
Correct Answer: E
Explanation: In Azure Machine Learning Studio, a custom component for use in a designer pipeline must be defined and registered first. This is done by authoring and uploading a component specification in a .yaml file. Once the YAML is uploaded, the component becomes available in the designer. A datastore already exists by default in a workspace, and pipeline endpoints or linked services are not prerequisites for using a custom component in a designer pipeline.
Question 44
You manage an Azure Machine Learning workspace.
You must create and configure a compute cluster for a training job by using Python SDK v2.
You need to create a persistent Azure Machine Learning compute resource, specifying the fewest possible properties.
Which two properties should you define? Each correct answer presents part of the solution.
NOTE: Each correct selection is worth one point.
A. size
B. win_instances
C. type
D. name
E. max_instances
Show Answer
Correct Answer: A, E
Explanation: When creating a persistent Azure Machine Learning compute cluster with the Python SDK v2 (AmlCompute), Azure provides smart defaults for most settings. The minimal properties you need to explicitly define are the VM size and the maximum number of instances for autoscaling. Other aspects, such as the compute type, are implied by using AmlCompute, and remaining settings are optional.
Question 45
You manage an Azure Machine Learning workspace. You design a training job that is configured with a serverless compute.
The serverless compute must have a specific instance type and count.
You need to configure the serverless compute by using Azure Machine Learning Python SDK v2.
What should you do?
A. Specify the compute name by using the compute parameter of the command job.
B. Configure the tier parameter to Dedicated VM.
C. Initialize and specify the ResourceConfiguration class.
D. Initialize AmiCompute class with size and type specification.
Show Answer
Correct Answer: C
Explanation: For Azure Machine Learning Python SDK v2, serverless compute is configured at the job level using the ResourceConfiguration class. By setting instance_type and instance_count in ResourceConfiguration and assigning it to the job’s resources parameter, you control the VM size and number of instances for serverless execution. Other options relate to named computes or managed clusters, not serverless jobs.
Question 46
You manage an Azure Machine Learning workspace named proj1.
You plan to use assets defined in proj1 to create a pipeline in the Machine Learning studio designer.
You need to set the Registry name filter to display only the list of assets defined in proj1.
What should you set the Registry name filter to?
A. proj1
B. azureml-meta
C. azureml
D. workspace
Show Answer
Correct Answer: D
Explanation: In Azure Machine Learning Designer, assets that are created within the current Azure Machine Learning workspace are stored in the **Workspace registry**. Setting the Registry name filter to **workspace** limits the view to only those assets defined in that workspace (proj1), excluding shared or Microsoft-provided registries such as azureml.
Question 48
DRAG DROP
-
You manage an Azure Machine Learning workspace.
You plan to import and wrangle data stored in Azure Data Lake Storage Gen2 with Apache Spark.
You need to start interactive data wrangling with the user identity passthrough.
Which three actions should you perform in sequence? To answer, move the appropriate actions from the list of actions to the answer area and arrange them in the correct order.
Show Answer
Correct Answer: Assign the user identity to Contributor and Storage Blob Data Contributor roles.
Select Serverless Spark Compute from the Compute selection menu.
Format the URI as abfss://<FILE_SYSTEM_NAME>@<STORAGE_ACCOUNT_NAME>.dfs.core.windows.net/<PATH_TO_DATA>.
Explanation: User identity passthrough requires proper RBAC first, then an interactive Serverless Spark session, and ADLS Gen2 access uses the abfss URI scheme for Spark-based data wrangling.
Question 49
You have an Azure Machine Learning workspace.
You plan to use the workspace to set up automated machine learning training for an image classification model.
You need to choose the primary metric to optimize the model training.
Which primary metric should you choose?
A. r2_score
B. mean_absolute_error
C. iou
D. median_absolute_error
Show Answer
Correct Answer: C
Explanation: For automated machine learning on images, the metric must be appropriate for computer vision tasks. Among the options, r2_score, mean_absolute_error, and median_absolute_error are regression metrics and are not suitable for image models. Intersection over Union (IoU) is a standard evaluation metric used in vision workloads (such as object detection and segmentation) and is the only image-appropriate option provided, making it the correct choice from the given answers.
Question 50
You have an Azure Machine Learning workspace named WS1.
You plan to use Azure Machine Learning SDK v2 to register a model as an asset in WS1 from an artifact generated by an MLflow run. The artifact resides in a named output of a job used for the model training.
You need to identify the syntax of the path to reference the model when you register it.
Which syntax should you use?
A. t/ /model/
B. azureml://registries
C. mlflow-model/
D. azureml://jobs/
Show Answer
Correct Answer: D
Explanation: When registering a model in Azure Machine Learning SDK v2 from an artifact produced by a training job, the model path must reference the job output using the AzureML URI scheme. This uses the syntax azureml://jobs/<job-name>/outputs/<output-name>/artifacts/<path>. Therefore, the correct syntax prefix is azureml://jobs/.
Question 51
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 in this section, you will NOT be able to return to it. As a result, these questions will not appear in the review screen.
You have an Azure Machine Learning workspace.
You plan to tune model hyperparameters by using a sweep job.
You need to find a sampling method that supports early termination of low-performance jobs and continuous hyperparameters.
Solution: Use the grid sampling method over the hyperparameter space.
Does the solution meet the goal?
A. Yes
B. No
Show Answer
Correct Answer: B
Explanation: The goal requires a sampling method that supports both early termination of low‑performance jobs and continuous hyperparameters. In Azure Machine Learning, grid sampling can work with early termination policies, but it only supports discrete hyperparameter values, not continuous ranges. Therefore, the proposed solution does not meet the stated goal.
Question 52
DRAG DROP
-
You have an Azure Machine Learning workspace named WS1 and a GitHub account named account1 that hosts a private repository named repo1.
You need to clone repo1 to make it available directly from WS1. The configuration must maximize the performance of the repo1 clone.
Which four actions should you perform in sequence?
Show Answer
Correct Answer: Create a compute instance.
Open a terminal window.
Generate a Secure Shell (SSH) key pair.
Add a public key to account1.
Explanation: To maximize clone performance in Azure ML, use a compute instance and clone locally via SSH. The compute instance provides fast, persistent storage; the terminal enables git operations; generating an SSH key allows secure access; adding the public key to GitHub enables high‑performance SSH cloning of the private repo.
Question 53
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 in this section, you will NOT be able to return to it. As a result, these questions will not appear in the review screen.
You have an Azure Machine Learning workspace named Workspace1. Workspace1 has a registered MLflow model named model1 with PyFunc flavor.
You plan to deploy model1 to an online endpoint named endpoint1 without egress connectivity by using Azure Machine Learning Python SDK v2.
You have the following code:
You need to add a parameter to the ManagedOnlineDeployment object to ensure the model deploys successfully.
Solution: Add the scoring_script parameter.
Does the solution meet the goal?
A. Yes
B. No
Show Answer
Correct Answer: B
Explanation: Adding a scoring_script parameter is not required to deploy an MLflow PyFunc model to a managed online endpoint, even in a no-egress scenario. Azure ML supports no-code deployment for MLflow models, where the scoring script and environment are inferred automatically. Egress connectivity is addressed through model packaging or workspace/network configuration, not by specifying a scoring_script. Therefore, the proposed solution does not meet the goal.
$19
Get all 477 questions with detailed answers and explanations
Instant download HTML + PDF delivered the moment payment clears.
Secure Stripe checkout we never see or store your card details.
7-day refund if files are defective see our refund policy.
