Google Professional Machine Learning Engineer Exam - Topic 6 Question 82 Discussion

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

Question #: 82
Topic #: 6

[All Professional Machine Learning Engineer Questions]

You work as an ML engineer at a social media company, and you are developing a visual filter for users' profile photos. This requires you to train an ML model to detect bounding boxes around human faces. You want to use this filter in your company's iOS-based mobile phone application. You want to minimize code development and want the model to be optimized for inference on mobile phones. What should you do?

ATrain a model using AutoML Vision and use the ''export for Core ML'' option.

BTrain a model using AutoML Vision and use the ''export for Coral'' option.

CTrain a model using AutoML Vision and use the ''export for TensorFlow.js'' option.

DTrain a custom TensorFlow model and convert it to TensorFlow Lite (TFLite).

Show Suggested Answer

Suggested Answer: D

by Della at Jul 11, 2024, 01:48 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

Cheryll

6 days ago

I think D might be better for custom needs.

upvoted 0 times

...

Deonna

12 days ago

A is the best choice for iOS apps!

upvoted 0 times

...

Cordelia

18 days ago

I’m leaning towards training a custom TensorFlow model and converting it to TFLite, but I’m not confident about the export options.

upvoted 0 times

...

Van

23 days ago

I feel like I saw something about exporting for Coral in a study session, but I don't think that's relevant for iOS apps.

upvoted 0 times

...

Joesph

28 days ago

I remember practicing a question about model optimization for mobile, and I think TensorFlow Lite could be a good choice for that.

upvoted 0 times

...

Cortney

1 month ago

I think using AutoML Vision and exporting for Core ML makes sense since it's for iOS, but I'm not completely sure if that's the best option.

upvoted 0 times

...

Angelo

1 month ago

I like the idea of minimizing code development, so I'm inclined to go with option A. Exporting the AutoML Vision model for Core ML sounds like the most straightforward approach.

upvoted 0 times

...

Gerri

1 month ago

Option B, exporting for Coral, is interesting, but I don't think that's the best fit since we're targeting an iOS app, not an edge device. I'm leaning towards A or D.

upvoted 0 times

...

Kelvin

1 month ago

Hmm, I'm a bit unsure about this one. I'm wondering if option D, training a custom TensorFlow model and converting it to TFLite, might give me more control and flexibility. I'll have to think about the tradeoffs.

upvoted 0 times

...

Allene

1 month ago

I think I'll go with option A. Using AutoML Vision and exporting for Core ML seems like the easiest way to get a model optimized for iOS deployment.

upvoted 0 times

...

Erin

1 month ago

This looks like a tricky networking question. I'm not too familiar with the specifics of ONTAP cluster networking, so I'll need to think this through carefully.

upvoted 0 times

...

Fletcher

2 months ago

Alright, I've got a strategy for this. I'll go through each statement and try to recall which Oracle module or application typically owns each of the shared entities listed.

upvoted 0 times

...

Carli

2 months ago

Therapeutic substitution means different chemical entity, right? So we're talking about swapping drugs within the same class. I'm betting approval matters here.

upvoted 0 times

...

Glory

6 months ago

Wait, is this a trick question? I'm gonna play it safe and go with option A. Ain't nobody got time for custom TensorFlow models when you can just let the AI do the heavy lifting. Now, where's the nearest coffee shop?

upvoted 0 times

...

Crissy

6 months ago

Oooh, decisions, decisions. I'm gonna have to go with option D. Sure, it might take a bit more work, but think of the bragging rights! I'll be the envy of all the other ML engineers at the company picnic.

upvoted 0 times

Kanisha

4 months ago

Exactly! And we can ensure that the model is specifically tailored for inference on mobile phones.

upvoted 0 times

...

Chantell

5 months ago

Definitely! Plus, having that custom TensorFlow model will give us more control over the optimization process.

upvoted 0 times

...

Nan

5 months ago

I think option D is the way to go. It might be more work, but the results will be worth it!

upvoted 0 times

...

Albina

5 months ago

D) Train a custom TensorFlow model and convert it to TensorFlow Lite (TFLite).

upvoted 0 times

...

Stanton

5 months ago

C) Train a model using AutoML Vision and use the ''export for TensorFlow.js'' option.

upvoted 0 times

...

Colette

5 months ago

B) Train a model using AutoML Vision and use the ''export for Coral'' option.

upvoted 0 times

...

Louann

5 months ago

A) Train a model using AutoML Vision and use the ''export for Core ML'' option.

upvoted 0 times

...

Genevive

6 months ago

Alright, time to put on my tech-savvy hat. Option A seems like the most efficient choice here. I mean, who wants to get bogged down in the nitty-gritty of model training when you can just let AutoML do its thing?

upvoted 0 times

...

Yong

6 months ago

Hmm, I'm feeling like a rebel today. Option D sounds like the way to go - train a custom TensorFlow model and then convert it to TensorFlow Lite. Gotta keep those mobile users happy, am I right?

upvoted 0 times

Nohemi

5 months ago

Frederick: Absolutely, optimizing for mobile is crucial. Custom models can give us more control over the process.

upvoted 0 times

...

Frederick

5 months ago

User 2: Yeah, that's a solid approach. Keeping the mobile users in mind is key for a smooth experience.

upvoted 0 times

...

Caprice

5 months ago

User 1: Option D sounds like a good choice. Custom TensorFlow model converted to TensorFlow Lite for mobile optimization.

upvoted 0 times

...

Deangelo

7 months ago

Whoa, looks like we've got a real brainteaser here! I'd go with option A - keep it simple and let AutoML Vision do the heavy lifting. Plus, Core ML is optimized for iOS, so it's a no-brainer.

upvoted 0 times