Exam Certified Machine Learning Associate All Questions

View all questions & answers for the Certified Machine Learning Associate exam

Exam Certified Machine Learning Associate topic 1 question 8 discussion

Actual exam question from Databricks's Certified Machine Learning Associate

Question #: 8
Topic #: 1

[All Certified Machine Learning Associate Questions]

A data scientist has created two linear regression models. The first model uses price as a label variable and the second model uses log(price) as a label variable. When evaluating the RMSE of each model by comparing the label predictions to the actual price values, the data scientist notices that the RMSE for the second model is much larger than the RMSE of the first model.
Which of the following possible explanations for this difference is invalid?

A. The second model is much more accurate than the first model
B. The data scientist failed to exponentiate the predictions in the second model prior to computing the RMSE
C. The data scientist failed to take the log of the predictions in the first model prior to computing the RMSE
D. The first model is much more accurate than the second model
E. The RMSE is an invalid evaluation metric for regression problems

Show Suggested Answer

Suggested Answer: B 🗳️

by vikasgautam at May 18, 2024, 4:21 a.m.

Comments

Submit Cancel

AbhirupAz900

1 month ago

Selected Answer: E

RMSE (Root Mean Square Error) is a standard and valid evaluation metric for regression problems. It measures the average magnitude of error and is widely used.

upvoted 1 times

...

ovokpus

1 month, 1 week ago

Selected Answer: C

He is looking for an invalid explanation. B is valid, but C is invalid as an explanation. He ought to exponentiate the predictions of the second model, and leave the first model as is.

upvoted 1 times

...

piwardox

4 months, 3 weeks ago

Selected Answer: D

I think the correct one should be D! As stated in the exercise prompt, in the first model I considered the RMSE to be the RMSE between the price (target variable) and the predictions, BUT the second to be the RMSE between the log(price) (target variable) and the predictions. In that case, you don't need to exponentiate anything!

upvoted 1 times

piwardox

4 months, 3 weeks ago

Rethinking it for one minute, I was wrong. Because in the case I imagined, there should not be so much difference between the RMSEs

upvoted 1 times

...

Deuterium44

7 months, 1 week ago

A is the correct answer, A and D

upvoted 2 times

Deuterium44

7 months, 1 week ago

A and D are cannot be true in the same time, and the less the RMSD is, the more the model is accurate. Thus, as model 1 as a lower RMSD than model 2, the invalid proposition is A

upvoted 1 times

...

jaydip650

7 months, 3 weeks ago

. The second model is much more accurate than the first model This explanation seems contradictory because if the second model were more accurate, its RMSE would be expected to be smaller, not larger.

upvoted 1 times

...

mj_infomotion

8 months, 1 week ago

Selected Answer: B

Needs to bi expon due to the logarithmic transformation before

upvoted 1 times

...

SusanDeeg

8 months, 2 weeks ago

The question is Which of the following possible explanations for this difference is INVALID? It would have to be E since RMSE is used for regression frequently. That cannot be the explanation.

upvoted 3 times

...

Soumyashree

1 year ago

Selected Answer: B

The second model uses log(price) as the label variable. If the data scientist directly computes the RMSE on the predicted log values without exponentiating them back to the original price scale, the errors will be much larger, leading to a higher RMSE.

upvoted 2 times

...

vikasgautam

1 year, 1 month ago

Selected Answer: B

the current answer is wrong.. RMSE is used for regression all the time

upvoted 3 times

...