Exam AWS Certified Machine Learning - Specialty All Questions

View all questions & answers for the AWS Certified Machine Learning - Specialty exam

Exam AWS Certified Machine Learning - Specialty topic 1 question 228 discussion

Exam question from Amazon's AWS Certified Machine Learning - Specialty

Question #: 228
Topic #: 1

[All AWS Certified Machine Learning - Specialty Questions]

A retail company wants to use Amazon Forecast to predict daily stock levels of inventory. The cost of running out of items in stock is much higher for the company than the cost of having excess inventory. The company has millions of data samples for multiple years for thousands of items. The company’s purchasing department needs to predict demand for 30-day cycles for each item to ensure that restocking occurs.

A machine learning (ML) specialist wants to use item-related features such as "category," "brand," and "safety stock count." The ML specialist also wants to use a binary time series feature that has "promotion applied?" as its name. Future promotion information is available only for the next 5 days.

The ML specialist must choose an algorithm and an evaluation metric for a solution to produce prediction results that will maximize company profit.

Which solution will meet these requirements?

A. Train a model by using the Autoregressive Integrated Moving Average (ARIMA) algorithm. Evaluate the model by using the Weighted Quantile Loss (wQL) metric at 0.75 (P75).
B. Train a model by using the Autoregressive Integrated Moving Average (ARIMA) algorithm. Evaluate the model by using the Weighted Absolute Percentage Error (WAPE) metric.
C. Train a model by using the Convolutional Neural Network - Quantile Regression (CNN-QR) algorithm. Evaluate the model by using the Weighted Quantile Loss (wQL) metric at 0.75 (P75).
D. Train a model by using the Convolutional Neural Network - Quantile Regression (CNN-QR) algorithm. Evaluate the model by using the Weighted Absolute Percentage Error (WAPE) metric.

Show Suggested Answer

Suggested Answer: C 🗳️

by sevosevo at March 18, 2023, 1:09 p.m.

Disclaimers:

- ExamTopics website is not related to, affiliated with, endorsed or authorized by Amazon.
- Trademarks, certification & product names are used for reference only and belong to Amazon.

Comments

Submit Cancel

blanco750

Highly Voted 1 year, 3 months ago

Selected Answer: C

WQL is particularly useful when there are different costs for underpredicting and overpredicting. By setting the weight (τ) of the wQL function, you can automatically incorporate differing penalties for underpredicting and overpredicting.

upvoted 6 times

...

loict

Highly Voted 9 months, 1 week ago

Selected Answer: C

A. NO - lot of data is available, better to use CNN-QR B. NO - lot of data is available, better to use CNN-QR C. YES - wQL is particularly useful when there are different costs for underpredicting and overpredicting (https://docs.aws.amazon.com/forecast/latest/dg/metrics.html#metrics-wQL) D. NO - WAPE will measure deviation, but no over vs. under forecasting

upvoted 5 times

...

Ganshank

Most Recent 1 month, 1 week ago

Selected Answer: A

The company has millions of data samples for multiple years for thousands of items. However, only 5 days worth of data is available for future promotions, therefore CNN-QR may not be the right fit as it requires large datasets. Also WQL is useful when there are different costs for underpredicting vs overpredicting. Therefore A seems to be the better option, than C.

upvoted 1 times

...

chet100

9 months, 3 weeks ago

It should be A I think. CNN is for image analysis

upvoted 1 times

...

Mickey321

10 months ago

Selected Answer: C

Option C also suggests evaluating the model by using the Weighted Quantile Loss (wQL) metric at 0.75 (P75). This metric measures the accuracy of a model at a specified quantile, which is a point in the distribution of possible outcomes2. For example, P75 means that 75% of the outcomes are below that point, and 25% are above it. This metric is suitable for your use case because it can incorporate different costs for underpredicting and overpredicting2. Since the cost of running out of items in stock is much higher for your company than the cost of having excess inventory, you can set a high weight (τ) for the wQL function to penalize underpredictions more than overpredictions2. This way, you can optimize your model to produce prediction results that will maximize your company profit.

upvoted 3 times

...

mawsman

1 year, 2 months ago

Selected Answer: A

upvoted 1 times

DimLam

7 months, 4 weeks ago

So, what is an argument for A here?

upvoted 1 times

...

austinoy

1 year, 3 months ago

I'll go with, C? https://docs.aws.amazon.com/forecast/latest/dg/aws-forecast-algo-cnnqr.html https://docs.aws.amazon.com/forecast/latest/dg/metrics.html#metrics-wQL

upvoted 3 times

...

sevosevo

1 year, 3 months ago

https://docs.aws.amazon.com/forecast/latest/dg/metrics.html https://docs.aws.amazon.com/forecast/latest/dg/aws-forecast-recipe-arima.html https://docs.aws.amazon.com/forecast/latest/dg/aws-forecast-algo-cnnqr.html

upvoted 1 times

...