Google Professional Machine Learning Engineer Exam - Topic 9 Question 36 Discussion

Actual exam question for Google's Professional Machine Learning Engineer exam

Question #: 36
Topic #: 9

[All Professional Machine Learning Engineer Questions]

As the lead ML Engineer for your company, you are responsible for building ML models to digitize scanned customer forms. You have developed a TensorFlow model that converts the scanned images into text and stores them in Cloud Storage. You need to use your ML model on the aggregated data collected at the end of each day with minimal manual intervention. What should you do?

AUse the batch prediction functionality of Al Platform

BCreate a serving pipeline in Compute Engine for prediction

CUse Cloud Functions for prediction each time a new data point is ingested

DDeploy the model on Al Platform and create a version of it for online inference.

Show Suggested Answer

Suggested Answer: D

by Rebecka at May 04, 2022, 11:20 PM

Limited Time Offer

25%

Off

Get Premium Professional Machine Learning Engineer Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Quentin

4 months ago

Wait, are we sure A is the most efficient?

upvoted 0 times

...

Garry

4 months ago

D seems like a solid option for real-time needs!

upvoted 0 times

...

Deandrea

4 months ago

C sounds interesting, but isn't that too manual?

upvoted 0 times

...

Sophia

4 months ago

I think B could work too, but it might be overkill.

upvoted 0 times

...

Coral

5 months ago

A is the best choice for batch processing!

upvoted 0 times

...

Carin

5 months ago

Deploying the model for online inference sounds interesting, but I wonder if it’s necessary for daily batch processing. I think batch prediction might be simpler.

upvoted 0 times

...

Sommer

5 months ago

I feel like creating a serving pipeline in Compute Engine could work, but it might require more setup than just using the AI Platform's batch prediction.

upvoted 0 times

...

Myra

5 months ago

I'm not entirely sure, but I think using Cloud Functions might be more suited for real-time predictions rather than batch processing.

upvoted 0 times

...

Karl

5 months ago

I remember we discussed batch prediction in our last study session. It seems like a good fit for processing daily aggregated data without much manual work.

upvoted 0 times

...

Johana

5 months ago

Hmm, I'm a bit confused on this one. I'm not sure if the purpose is to underpin the prioritization of requirements or to position DSDM projects for a successful outcome. I'll have to think this through more carefully.

upvoted 0 times

...

Walker

5 months ago

Okay, I think the key here is to focus on accelerating the IP ramp-up process. If they can migrate to pre-warmed IP addresses, that could help them get up and running faster.

upvoted 0 times

...