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

Amazon MLA-C01 Exam - Topic 1 Question 25 Discussion

Posted by Cecil on Sep 29, 2026 at 04:33 PM

An ML engineer normalized training data by using min-max normalization in AWS Glue DataBrew. The ML engineer must normalize production inference data in the same way before passing the data to the model.

Which solution will meet this requirement?

Suggested Answer: B

AWS ML best practices state that data preprocessing applied during training must be applied identically during inference. For min-max normalization, this requires reusing the minimum and maximum values calculated from the training dataset.

If production data is normalized using different statistics, the feature distributions will differ from what the model learned, leading to degraded prediction accuracy. AWS documentation explicitly warns against recomputing normalization parameters on inference data.

Options A, C, and D introduce data leakage or inconsistent feature scaling. Option B ensures consistency between training and inference pipelines and preserves model integrity.

Therefore, Option B is the correct and AWS-aligned solution.


Get Premium MLA-C01 Questions as Interactive Web-Based Practice Test or PDF Real Exam Format Detailed Explanations Instant Access Get Your Practice Test
13 comments · newest first
KA
Karen Wilson 6 days ago
Maybe B is not correct because the tricky part is that preprocessing must be identical at inference to training.
JI
Jiyeon Huang 22 days ago
Going with C since recomputing min and max from production data breaks the tricky requirement of matching training scaling.
AL
Alejandro Diaz 7 days ago
↳ replying to Jiyeon Huang
Same here, the requirement rules out the others.
JU
Julien Bruno 18 days ago
↳ replying to Jiyeon Huang
This matches what I got, and recomputing per batch would still shift feature scales between batches.
MA
Margot 3 days ago
B is definitely the best option here!
GW
Gwenn 8 days ago
D? Really? That seems inefficient.
LU
Luis 13 days ago
C sounds reasonable, but it might not match training data.
MU
Murray 19 days ago
I think A could work too, but it’s risky.
AM
Amie 24 days ago
B is the way to go! Consistency is key.
CL
Claribel 29 days ago
I vaguely recall that using statistics from the training set is crucial, so I lean towards option B, but I wonder if there are any exceptions to that rule.
DI
Dick 1 month ago
I'm a bit confused about option C. It seems like recalculating min-max stats from production samples could lead to inconsistencies, right?
FE
Felix 1 month ago
I remember a similar question where we had to maintain consistency between training and production data. I feel like option B is the right choice here too.
MA
Marg 1 month ago
I think option B makes the most sense since it uses the same statistics from the training set, but I'm not completely sure if that's the best approach.
0/2000 characters

Save Cancel