Google Professional Data Engineer Exam - Topic 3 Question 31 Discussion

Actual exam question for Google's Professional Data Engineer exam

Question #: 31
Topic #: 3

[All Professional Data Engineer Questions]

You work for a manufacturing company that sources up to 750 different components, each from a different supplier. You've collected a labeled dataset that has on average 1000 examples for each unique component. Your team wants to implement an app to help warehouse workers recognize incoming components based on a photo of the component. You want to implement the first working version of this app (as Proof-Of-Concept) within a few working days. What should you do?

AUse Cloud Vision AutoML with the existing dataset.

BUse Cloud Vision AutoML, but reduce your dataset twice.

CUse Cloud Vision API by providing custom labels as recognition hints.

DTrain your own image recognition model leveraging transfer learning techniques.

Show Suggested Answer

Suggested Answer: A

by Jettie at May 05, 2022, 09:05 PM

Limited Time Offer

25%

Off

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

Contribute your Thoughts:

Submit Cancel

Catarina

4 months ago

Wait, can we really get it done in just a few days?

upvoted 0 times

...

Regenia

4 months ago

C could work too, especially if we need it fast.

upvoted 0 times

...

Ines

4 months ago

D sounds cool, but isn't it more time-consuming?

upvoted 0 times

...

Temeka

4 months ago

I think reducing the dataset (B) might hurt accuracy.

upvoted 0 times

...

Ceola

5 months ago

A is the quickest way to get started!

upvoted 0 times

...

Ozell

5 months ago

Training our own model with transfer learning sounds appealing, but I worry it might take longer than a few days.

upvoted 0 times

...

Keena

5 months ago

I think option C could work too, especially if we provide good custom labels. It might save time.

upvoted 0 times

...

Jesusita

5 months ago

I'm not entirely sure, but reducing the dataset in option B might not be the best idea since we have enough examples.

upvoted 0 times

...

Ronald

5 months ago

I remember we discussed using Cloud Vision AutoML for similar projects, so option A seems like a solid choice.

upvoted 0 times

...

Francisca

5 months ago

This looks like a tricky question, but I think I can narrow it down. SD-Access is a Cisco solution, so the answer is likely one of their products.

upvoted 0 times

...

Salina

5 months ago

Alright, let's do this. I just need to remember the right formulas and unit conversions. As long as I take my time and double-check my work, I should be able to nail this question.

upvoted 0 times

...