Google Professional Machine Learning Engineer Exam - Topic 7 Question 20 Discussion

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

Question #: 20
Topic #: 7

[All Professional Machine Learning Engineer Questions]

You need to train a computer vision model that predicts the type of government ID present in a given image using a GPU-powered virtual machine on Compute Engine. You use the following parameters:

* Optimizer: SGD

* Image shape = 224x224

* Batch size = 64

* Epochs = 10

* Verbose = 2

During training you encounter the following error: ResourceExhaustedError: out of Memory (oom) when allocating tensor. What should you do?

AChange the optimizer

BReduce the batch size

CChange the learning rate

DReduce the image shape

Show Suggested Answer

Suggested Answer: B

by Celeste at May 04, 2022, 01:46 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

Stanton

4 months ago

Not sure if just lowering the batch size is enough...

upvoted 0 times

...

Erinn

4 months ago

Wait, can reducing image size really solve this?

upvoted 0 times

...

Stacey

4 months ago

Changing the optimizer won't help with memory issues.

upvoted 0 times

...

Tenesha

4 months ago

I agree, batch size affects memory usage a lot.

upvoted 0 times

...

Lindsey

5 months ago

Reducing the batch size is usually a good first step.

upvoted 0 times

...

Jestine

5 months ago

I recall that adjusting the learning rate usually affects convergence speed, but I'm not convinced it would help with an out of memory error.

upvoted 0 times

...

Loreta

5 months ago

I practiced a similar question where reducing the image size helped with memory constraints. Maybe that could work here too?

upvoted 0 times

...

Reuben

5 months ago

I remember we discussed how reducing the batch size can help with memory issues during training. It seems like a good option here.

upvoted 0 times

...

Dyan

5 months ago

I'm not entirely sure, but I think changing the optimizer might not directly solve the out of memory error. It feels like a workaround rather than a fix.

upvoted 0 times

...

Valentine

5 months ago

Hmm, this looks like a tricky one. I'll need to think carefully about the capabilities of Automation script settings.

upvoted 0 times

...

Adolph

5 months ago

Hmm, this looks tricky. I'll need to carefully review the options and think through the potential causes.

upvoted 0 times

...

Chantell

5 months ago

I'm a bit confused about the default action of an IP filter. Is it to forward or drop packets? I'll need to double-check that detail.

upvoted 0 times

...

Roxane

5 months ago

I've got this! The X part represents the major version of MySQL. That's the most important part of the version number, so it makes sense that it would be denoted by the X.

upvoted 0 times

...