Google Professional Machine Learning Engineer Exam - Topic 5 Question 35 Discussion

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

Question #: 35
Topic #: 5

[All Professional Machine Learning Engineer Questions]

You work for a global footwear retailer and need to predict when an item will be out of stock based on historical inventory dat

a. Customer behavior is highly dynamic since footwear demand is influenced by many different factors. You want to serve models that are trained on all available data, but track your performance on specific subsets of data before pushing to production. What is the most streamlined and reliable way to perform this validation?

AUse the TFX ModelValidator tools to specify performance metrics for production readiness

BUse k-fold cross-validation as a validation strategy to ensure that your model is ready for production.

CUse the last relevant week of data as a validation set to ensure that your model is performing accurately on current data

DUse the entire dataset and treat the area under the receiver operating characteristics curve (AUC ROC) as the main metric.

Show Suggested Answer

Suggested Answer: A

by Antione at May 06, 2022, 08:21 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

Jacinta

4 months ago

Wait, treating the whole dataset as one? Isn’t that a bit too broad?

upvoted 0 times

...

Bulah

4 months ago

Totally agree, A is streamlined and reliable!

upvoted 0 times

...

Dominic

4 months ago

Using just the last week of data? Seems risky, what if demand spikes?

upvoted 0 times

...

Rashad

4 months ago

K-fold cross-validation is solid too, but it might take longer.

upvoted 0 times

...

Tina

5 months ago

I think option A is the best choice for production readiness.

upvoted 0 times

...

Murray

5 months ago

I feel like using the entire dataset and focusing on AUC ROC (option D) might overlook the need for targeted validation, but I’m not completely confident about that.

upvoted 0 times

...

Marquetta

5 months ago

I practiced using the last week of data for validation in a similar question, so option C could be a good way to ensure accuracy with current trends.

upvoted 0 times

...

Art

5 months ago

I’m not entirely sure, but I think k-fold cross-validation (option B) is a common method for validating models. It might help with the dynamic nature of customer behavior.

upvoted 0 times

...

Darrin

5 months ago

I remember discussing the importance of using specific performance metrics for production readiness, so option A seems like a solid choice.

upvoted 0 times

...

Judy

5 months ago

Ah, I remember learning about this in class. The key is to identify the interface types that can act as the endpoints for a tunnel. Time to put my networking knowledge to the test!

upvoted 0 times

...

Flo

5 months ago

I'm a bit unsure about this one. The options seem to be describing different types of information processing systems, but I'm not sure which one is the most relevant to the question. I'll need to think it through carefully.

upvoted 0 times

...

Ernie

5 months ago

Okay, let me think this through. The question is asking about sharing a folder on a Windows 10 workgroup computer, and making it available to offline users. The public share option doesn't seem quite right for that scenario. I think I'll go with "HomeGroup" as the best answer.

upvoted 0 times

...

Cherry

5 months ago

I'm not entirely sure, but I feel like it might also be linked to avoiding actions that discredit the profession. Isn't that what option B suggests?

upvoted 0 times

...

Avery

5 months ago

Hmm, I'm a bit unsure about this one. I'm thinking option A with Terraform might be the way to go, but I'm not sure if that would be the fastest approach. I'll have to think this through a bit more.

upvoted 0 times

...