HOTSPOT
-
You use an Azure Machine Learning workspace. The default datastore contains comma-separated values (CSV) files.
The CSV files must be made available for use in experiments and data processing pipelines. The files must be loaded directly into pandas dataframes.
How should you complete the code? To answer, select the appropriate options in the answer area.
NOTE: Each correct selection is worth one point.
Explanation: CSV files in a datastore should be created as a TabularDataset so they can be loaded directly into pandas DataFrames. This is done using the Dataset class with Tabular.from_delimited_files, which is designed for structured, delimited text files like CSV.
Question 180
HOTSPOT
-
You create an Azure Machine Learning workspace. You use the Azure Machine Learning SDK for Python.
You must create a dataset from remote paths. The dataset must be reusable within the workspace.
You need to create the dataset.
How should you complete the following code segment? To answer, select the appropriate options in the answer area.
NOTE: Each correct selection is worth one point.
Explanation: The source files are CSVs accessed via HTTPS, so a tabular dataset must be created from delimited files. To make the dataset reusable within the Azure ML workspace, it must be registered using the register() method.
Question 181
You are developing a machine learning model.
You must inference the machine learning model for testing.
You need to use a minimal cost compute target.
Which two compute targets should you use? Each correct answer presents a complete solution.
NOTE: Each correct selection is worth one point.
A. Azure Machine Learning Kubernetes
B. Azure Databricks
C. Remote VM
D. Local web service
E. Azure Container Instances
Show Answer
Correct Answer: D, E
Explanation: For inference during testing, the goal is to minimize cost rather than provide scalable or production-grade hosting. A local web service runs on the developer’s machine and incurs no Azure compute cost, making it the cheapest option. Azure Container Instances (ACI) is a serverless, lightweight compute target that spins up on demand and charges only for usage, making it the lowest-cost Azure-managed option for test inference. Other options like AKS, Databricks, or remote VMs are designed for production or always-on workloads and are more expensive.
Question 182
You use the Azure Machine Learning SDK for Python to create a pipeline that includes the following step:
The output of the step run must be cached and reused on subsequent runs when the source_directory value has not changed.
You need to define the step.
What should you include in the step definition?
A. allow_reuse
B. version
C. data.as_input(name=…)
D. hash_paths
Show Answer
Correct Answer: A
Explanation: To ensure that a pipeline step’s output is cached and reused on subsequent runs when nothing relevant has changed (including the source_directory), the step must have reuse enabled. The allow_reuse parameter controls this behavior. When allow_reuse=True (the default), Azure Machine Learning reuses the results from a previous run if the step’s inputs, parameters, and source code remain unchanged, avoiding re-execution.
Question 183
HOTSPOT
-
You have a binary classifier that predicts positive cases of diabetes within two separate age groups.
The classifier exhibits a high degree of disparity between the age groups.
You need to modify the output of the classifier to maximize its degree of fairness across the age groups and meet the following requirements:
• Eliminate the need to retrain the model on which the classifier is based.
• Minimize the disparity between true positive rates and false positive rates across age groups.
Which algorithm and parity constraint should you use? To answer, select the appropriate options in the answer area.
NOTE: Each correct selection is worth one point.
Explanation: Post-processing is required because the model must not be retrained. The Threshold Optimizer adjusts decision thresholds per group after training. Equalized odds minimizes disparity in both true positive rates and false positive rates across age groups, directly matching the fairness requirement.
Question 184
DRAG DROP
-
You have an Azure Machine Learning workspace that contains a training cluster and an inference cluster.
You plan to create a classification model by using the Azure Machine Learning designer.
You need to ensure that client applications can submit data as HTTP requests and receive predictions as responses.
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: Create a pipeline that trains a classification model and run the pipeline on the compute cluster.
Create a real-time inference pipeline and run the pipeline on the compute cluster.
Deploy a service to the inference cluster.
Explanation: First, a training pipeline is required to produce a trained classification model. Next, a real-time inference pipeline is created to enable online predictions. Finally, the inference pipeline is deployed as a service to the inference cluster so client applications can send HTTP requests and receive prediction responses.
Question 185
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 create an Azure Machine Learning pipeline named pipeline1 with two steps that contain Python scripts. Data processed by the first step is passed to the second step.
You must update the content of the downstream data source of pipeline1 and run the pipeline again.
You need to ensure the new run of pipeline1 fully processes the updated content.
Solution: Change the value of the compute_target parameter of the PythonScriptStep object in the two steps.
Does the solution meet the goal?
A. Yes
B. No
Show Answer
Correct Answer: B
Explanation: Changing the compute_target only alters where the step runs and is not the correct or intended way to ensure updated data is processed. To guarantee reprocessing of updated downstream data, you should change the input data (for example, update the dataset/version) or disable step reuse. Therefore, the proposed solution does not meet the goal.
Question 186
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 create an Azure Machine Learning pipeline named pipeline1 with two steps that contain Python scripts. Data processed by the first step is passed to the second step.
You must update the content of the downstream data source of pipeline1 and run the pipeline again.
You need to ensure the new run of pipeline1 fully processes the updated content.
Solution: Set the regenerate_outputs parameter of the pipeline1 experiment’s run submit method to True.
Does the solution meet the goal?
A. Yes
B. No
Show Answer
Correct Answer: A
Explanation: Setting regenerate_outputs=True forces all pipeline steps to rerun and prevents reuse of cached outputs from previous runs. This ensures that any updated downstream data is fully reprocessed when the pipeline is run again, which meets the stated goal.
Question 187
You create an MLflow model.
You must deploy the model to Azure Machine Learning for batch inference.
You need to create the batch deployment.
Which two components should you use? Each correct answer presents a complete solution.
NOTE: Each correct selection is worth one point.
A. Environment
B. Model files
C. Online endpoint
D. Kubernetes online endpoint
E. Compute target
Show Answer
Correct Answer: B, E
Explanation: For batch inference with an MLflow model in Azure Machine Learning, you need the registered model (model files) and a compute target to run the batch job. MLflow models already encapsulate the environment and scoring logic, so you do not need to explicitly define an environment or endpoint type.
Question 188
You train and publish a machine learning model.
You need to run a pipeline that retrains the model based on a trigger from an external system.
What should you configure?
A. Azure Data Catalog
B. Azure Batch
C. Azure Logic App
Show Answer
Correct Answer: C
Explanation: To retrain a published machine learning model based on a trigger from an external system, you need an event-driven orchestration service. Azure Logic Apps can listen for external events (such as HTTP requests, Event Grid events, or webhooks) and trigger Azure Machine Learning pipelines accordingly. Azure Data Catalog is for data discovery, and Azure Batch is for large-scale parallel compute, not event-based pipeline triggering.
$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.
