Google Professional Machine Learning Engineer Exam - Topic 9 Question 43 Discussion

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

Question #: 43
Topic #: 9

[All Professional Machine Learning Engineer Questions]

You have trained a model on a dataset that required computationally expensive preprocessing operations. You need to execute the same preprocessing at prediction time. You deployed the model on Al Platform for high-throughput online prediction. Which architecture should you use?

A* Validate the accuracy of the model that you trained on preprocessed data
* Create a new model that uses the raw data and is available in real time
* Deploy the new model onto Al Platform for online prediction

B* Send incoming prediction requests to a Pub/Sub topic
* Transform the incoming data using a Dataflow job
* Submit a prediction request to Al Platform using the transformed data
* Write the predictions to an outbound Pub/Sub queue

C* Stream incoming prediction request data into Cloud Spanner
* Create a view to abstract your preprocessing logic.
* Query the view every second for new records
* Submit a prediction request to Al Platform using the transformed data
* Write the predictions to an outbound Pub/Sub queue.

D* Send incoming prediction requests to a Pub/Sub topic
* Set up a Cloud Function that is triggered when messages are published to the Pub/Sub topic.
* Implement your preprocessing logic in the Cloud Function
* Submit a prediction request to Al Platform using the transformed data
* Write the predictions to an outbound Pub/Sub queue

Show Suggested Answer

Suggested Answer: D

by Gladys at May 08, 2022, 01:06 AM

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

Shalon

4 months ago

A new model for raw data? Sounds like a lot of extra work!

upvoted 0 times

...

Frederic

4 months ago

I’m leaning towards C, but it seems a bit complex for just preprocessing.

upvoted 0 times

...

Lyda

4 months ago

Wait, can we really handle all that data in real-time with option D?

upvoted 0 times

...

Emilio

4 months ago

Definitely agree with B, it’s efficient and scalable!

upvoted 0 times

...

Jordan

5 months ago

I think option B is the best choice for handling high-throughput predictions.

upvoted 0 times

...

Macy

5 months ago

I recall a similar question where we had to choose between streaming and batch processing. I wonder if that applies here as well.

upvoted 0 times

...

Nelida

5 months ago

I feel like option D could be a good choice too, especially with the Cloud Function handling preprocessing. But I’m not entirely confident.

upvoted 0 times

...

Delsie

5 months ago

I remember we discussed the importance of preprocessing at prediction time, but I'm not sure which architecture handles that best.

upvoted 0 times

...

Myra

5 months ago

I think option B sounds familiar because it involves using Dataflow for transformations, which we practiced in class.

upvoted 0 times

...

Silva

5 months ago

Hmm, I'm not sure about this one. I'll have to think it through carefully.

upvoted 0 times

...

Herminia

5 months ago

Hmm, this looks like a tricky one. I'll need to think through the confusion matrix and population event probability to figure out the sensitivity.

upvoted 0 times

...

Lashon

5 months ago

This seems straightforward, I'll focus on the key details about the new ONTAP cluster and automatic port placement.

upvoted 0 times

...

Cordie

5 months ago

Okay, let me think this through. The bug was fixed in the current 10.0.2.1 version, so the new version should increment the patch level, which would be 10.0.2.2. I'll go with option B.

upvoted 0 times

...

Cammy

5 months ago

Okay, I've seen this type of question before. I think the key is to identify the mandatory parameters for controlling access.

upvoted 0 times

...

Aleta

5 months ago

This one seems straightforward. I'll need to associate both the HTTP and HTTPS profiles to handle the HTTPS traffic and parse the headers.

upvoted 0 times

...