Microsoft DP-100 Exam - Topic 10 Question 46 Discussion

Actual exam question for Microsoft's DP-100 exam

Question #: 46
Topic #: 10

You use the Azure Machine Learning designer to create and run a training pipeline.

The pipeline must be run every night to inference predictions from a large volume of files. The folder where the files will be stored is defined as a dataset.

You need to publish the pipeline as a REST service that can be used for the nightly inferencing run.

What should you do?

ACreate a batch inference pipeline

BSet the compute target for the pipeline to an inference cluster

CCreate a real-time inference pipeline

DClone the pipeline

Show Suggested Answer

Suggested Answer: A

Azure Machine Learning Batch Inference targets large inference jobs that are not time-sensitive. Batch Inference provides cost-effective inference compute scaling, with unparalleled throughput for asynchronous applications. It is optimized for high-throughput, fire-and-forget inference over large collections of data.

You can submit a batch inference job by pipeline_run, or through REST calls with a published pipeline.

https://github.com/Azure/MachineLearningNotebooks/blob/master/how-to-use-azureml/machine-learning-pipelines/parallel-run/README.md

by Billye at May 06, 2022, 05:32 PM

Limited Time Offer

25%

Off

Get Premium DP-100 Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Christiane

4 months ago

I thought real-time pipelines were for immediate predictions, not nightly ones?

upvoted 0 times

...

Therese

4 months ago

Setting the compute target is important, but it’s not the main step here.

upvoted 0 times

...

Elouise

4 months ago

Wait, can you really publish a batch pipeline as a REST service?

upvoted 0 times

...

Olene

5 months ago

I agree, option A makes the most sense here!

upvoted 0 times

...

Mose

5 months ago

A batch inference pipeline is definitely the way to go for nightly runs.

upvoted 0 times

...

Sylvie

5 months ago

Cloning the pipeline seems like an option, but I don't see how that helps with the REST service for inferencing.

upvoted 0 times

...

Rima

5 months ago

I feel like this question is similar to one we practiced about real-time vs batch processing. I think batch inference makes more sense for nightly runs.

upvoted 0 times

...

Cruz

5 months ago

I'm not entirely sure, but I remember something about setting the compute target for the pipeline. Could that be relevant here?

upvoted 0 times

...

Hillary

5 months ago

I think we need to create a batch inference pipeline since we're dealing with a large volume of files every night.

upvoted 0 times

...

Lacey

5 months ago

I think the Cumulative Promotion is the best fit here. The company is looking to reward customers for making multiple purchases within a single month, which aligns with the description of that promotion type.

upvoted 0 times

...

Dick

5 months ago

I remember a question like this on practice tests! I think it was about how delivery is part of the sequence when an attachment is sent.

upvoted 0 times

...

Laurene

5 months ago

Wait, I'm confused. If the sessions are load-balanced, won't changing the Drain mode just shift the load to the other hosts? I'm not sure this is the right approach.

upvoted 0 times

...