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 30 Discussion

You have a demand forecasting pipeline in production that uses Dataflow to preprocess raw data prior to model training and prediction. During preprocessing, you employ Z-score normalization on data stored in BigQuery and write it back to BigQuery. New training data is added every week. You want to make the process more efficient by minimizing computation time and manual intervention. What should you do?
B) Translate the normalization algorithm into SQL for use with BigQuery
A) Normalize the data using Google Kubernetes Engine
C) Use the normalizer_fn argument in TensorFlow's Feature Column API
D) Normalize the data with Apache Spark using the Dataproc connector for BigQuery

Google Professional Machine Learning Engineer Exam - Topic 2 Question 30 Discussion

Actual exam question for Google's Professional Machine Learning Engineer exam
Question #: 30
Topic #: 2
[All Professional Machine Learning Engineer Questions]

You have a demand forecasting pipeline in production that uses Dataflow to preprocess raw data prior to model training and prediction. During preprocessing, you employ Z-score normalization on data stored in BigQuery and write it back to BigQuery. New training data is added every week. You want to make the process more efficient by minimizing computation time and manual intervention. What should you do?

Show Suggested Answer Hide Answer
Suggested Answer: B

by Lawrence at May 09, 2022, 05:33 AM

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
Cordelia
10 months ago
C is cool, but isn't it more for model training than preprocessing?
upvoted 0 times
...
Mammie
10 months ago
D might be overkill for just normalization, right?
upvoted 0 times
...
Clare
11 months ago
Wait, can you really do normalization in SQL? Sounds interesting!
upvoted 0 times
...
Lucille
11 months ago
I disagree, A could offer more flexibility with Kubernetes.
upvoted 0 times
...
Georgeanna
11 months ago
B seems like a solid choice, keeps it all in BigQuery.
upvoted 0 times
...
Yaeko
11 months ago
I feel like normalizing in Kubernetes could add unnecessary complexity, but I can't recall the exact reasons why.
upvoted 0 times
...
Micah
11 months ago
I practiced a similar question where using Apache Spark with Dataproc was highlighted as a way to handle large datasets efficiently.
upvoted 0 times
...
Dorsey
11 months ago
I'm not entirely sure, but I think using TensorFlow's Feature Column API might be more suited for model training rather than preprocessing in this case.
upvoted 0 times
...
Vanesa
11 months ago
I remember we discussed how using SQL for normalization could streamline the process since it's directly integrated with BigQuery.
upvoted 0 times
...
Latosha
11 months ago
Wait, is this asking about paraphrasing in a customer service context? I was thinking more generally. Let me re-read the question and options carefully to make sure I'm on the right track.
upvoted 0 times
...
Dacia
11 months ago
Implementing Windows AutoPilot and deploying policies through Intune seem like the most efficient ways to standardize the device deployments and minimize costs. I'll focus on those two options.
upvoted 0 times
...
Micah
11 months ago
Okay, let me see. The Session layer is all about managing the communication session, so an application-level firewall that inspects the actual application data seems like the right choice here. I'm feeling good about B.
upvoted 0 times
...

Save Cancel