Deal of The Day! Hurry Up, Grab the Special Discount - Save 25% - Ends In 00:00:00 Coupon code: SAVE25
Welcome to Pass4Success

- Free Preparation Discussions
Mail Us support@pass4success.com
Location US
Mob_nav_barsMENU

Google Professional Machine Learning Engineer Exam - Topic 2 Question 27 Discussion

Your team needs to build a model that predicts whether images contain a driver's license, passport, or credit card. The data engineering team already built the pipeline and generated a dataset composed of 10,000 images with driver's licenses, 1,000 images with passports, and 1,000 images with credit cards. You now have to train a model with the following label map: ['driversjicense', 'passport', 'credit_card']. Which loss function should you use?
D) Sparse categorical cross-entropy
A) Categorical hinge
B) Binary cross-entropy
C) Categorical cross-entropy

Google Professional Machine Learning Engineer Exam - Topic 2 Question 27 Discussion

Actual exam question for Google's Professional Machine Learning Engineer exam
Question #: 27
Topic #: 2
[All Professional Machine Learning Engineer Questions]

Your team needs to build a model that predicts whether images contain a driver's license, passport, or credit card. The data engineering team already built the pipeline and generated a dataset composed of 10,000 images with driver's licenses, 1,000 images with passports, and 1,000 images with credit cards. You now have to train a model with the following label map: ['driversjicense', 'passport', 'credit_card']. Which loss function should you use?

Show Suggested Answer Hide Answer
Suggested Answer: D

se sparse_categorical_crossentropy. Examples for above 3-class classification problem: [1] , [2], [3]


by Hermila at May 06, 2022, 02:50 PM

Limited Time Offer

25%

Off

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

Contribute your Thoughts:

0/2000 characters
Submit Cancel
Paulene
7 months ago
Totally agree with Categorical cross-entropy! Makes sense for this dataset.
upvoted 0 times
...
Oliva
7 months ago
Wait, why not binary cross-entropy? Isn’t that simpler?
upvoted 0 times
...
Adelina
8 months ago
Definitely Categorical cross-entropy! It fits the label map perfectly.
upvoted 0 times
...
Iesha
8 months ago
I think sparse categorical cross-entropy could work too, but not sure.
upvoted 0 times
...
Tammi
8 months ago
Categorical cross-entropy is the way to go for multi-class classification.
upvoted 0 times
...
Lilli
8 months ago
I feel like binary cross-entropy is for two classes, so it can't be the answer, but I'm not clear on the differences between the others.
upvoted 0 times
...
Walker
8 months ago
Categorical cross-entropy seems right because we have one-hot encoded labels, but I wonder if the imbalance in classes affects the choice.
upvoted 0 times
...
Jennie
8 months ago
I remember practicing with a similar question, and I think sparse categorical cross-entropy might be for when labels are integers, but I'm confused about the specifics here.
upvoted 0 times
...
Adolph
8 months ago
I think we should use categorical cross-entropy since we have multiple classes, but I'm not entirely sure.
upvoted 0 times
...
Linn
8 months ago
This is a tricky one. I know load balancing is important for scalability, but I'm not sure which specific pattern it's associated with. I'll have to make an educated guess and go with option C.
upvoted 0 times
...
Domonique
8 months ago
I'm torn between the variable options. If they go for a plain coupon swap at that rate, wouldn't it likely be LIBOR + 0.5% or even 1%?
upvoted 0 times
...
Royal
8 months ago
I'm a bit confused about the role of the risk committee. I feel like they have responsibilities, but I can't recall if it's exactly one of COSO's key principles.
upvoted 0 times
...
Cecil
8 months ago
Hmm, I'm a bit confused. I know 802.11b also operates at 2.4 GHz, but I'm not sure about the data rate. I'll have to think this through.
upvoted 0 times
...

Save Cancel