DP-100 Question #190: Real Exam Question with Answer & Explanation

Question

You create a batch inference pipeline by using the Azure ML SDK. You configure the pipeline parameters by executing the following code: You need to obtain the output from the pipeline execution. Where will you find the output?

Options

• Athe digit_identification.py script
• Bthe debug log
• Cthe Activity Log in the Azure portal for the Machine Learning workspace
• Dthe Inference Clusters tab in Machine Learning studio
• Ea file named parallel_run_step.txt located in the output folder

Explanation

{"question_number": 7, "correct_answer": "E", "explanation": "In Azure ML batch inference pipelines built with ParallelRunStep, prediction results from all parallel workers are automatically aggregated and written to a file named 'parallel_run_step.txt' inside the designated output folder. This is the fixed, standard output filename used by ParallelRunConfig - regardless of the input data or model used. After the pipeline run completes, you retrieve this file from the output folder (accessible via the pipeline run's output dataset or by downloading from the associated datastore path). The other options are incorrect: the digit_identification.py script is the scoring script, not an output location; the debug log contains diagnostic info; the Activity Log tracks Azure resource operations; and Inference Clusters manage real-time endpoints, not batch outputs.", "generated_by": "claude-sonnet", "llm_judge_score": 4}

No community discussion yet for this question.