Google examstrust google Professional-machine-learning-engineer Online Access by Malachi q116 vce pdf

Page: 17 / 21

Exam Name:	Google Professional Machine Learning Engineer
Exam Code:	Professional-Machine-Learning-Engineer Dumps
Vendor:	Google	Certification:	Machine Learning Engineer
Questions:	285 Q&A's	Shared By:	malachi

Question 68

You work for an online retailer. Your company has a few thousand short lifecycle products. Your company has five years of sales data stored in BigQuery. You have been asked to build a model that will make monthly sales predictions for each product. You want to use a solution that can be implemented quickly with minimal effort. What should you do?

Options:

Use Prophet on Vertex Al Training to build a custom model.

Use Vertex Al Forecast to build a NN-based model.

Use BigQuery ML to build a statistical AR1MA_PLUS model.

Use TensorFlow on Vertex Al Training to build a custom model.

Discussion

Question 69

You are building a real-time prediction engine that streams files which may contain Personally Identifiable Information (Pll) to Google Cloud. You want to use the Cloud Data Loss Prevention (DLP) API to scan the files. How should you ensure that the Pll is not accessible by unauthorized individuals?

Options:

Stream all files to Google CloudT and then write the data to BigQuery Periodically conduct a bulk scan of the table using the DLP API.

Stream all files to Google Cloud, and write batches of the data to BigQuery While the data is being written to BigQuery conduct a bulk scan of the data using the DLP API.

Create two buckets of data Sensitive and Non-sensitive Write all data to the Non-sensitive bucket Periodically conduct a bulk scan of that bucket using the DLP API, and move the sensitive data to the Sensitive bucket

Create three buckets of data: Quarantine, Sensitive, and Non-sensitive Write all data to the Quarantine bucket.

Periodically conduct a bulk scan of that bucket using the DLP API, and move the data to either the Sensitive or Non-Sensitive bucket

Discussion

Question 70

You built and manage a production system that is responsible for predicting sales numbers. Model accuracy is crucial, because the production model is required to keep up with market changes. Since being deployed to production, the model hasn't changed; however the accuracy of the model has steadily deteriorated. What issue is most likely causing the steady decline in model accuracy?

Options:

Poor data quality

Lack of model retraining

Too few layers in the model for capturing information

Incorrect data split ratio during model training, evaluation, validation, and test

Discussion

Answer:

Explanation:

Model retraining is the process of updating an existing machine learning model with new data and parameters to improve its performance and accuracy. Model retraining is essential for maintaining the relevance and validity of the model, especially when the data or the environment changes over time. Model retraining can help to avoid or reduce the effects of model degradation, which is the phenomenon of the model’s predictive performance decreasing as it is tested on new datasets within rapidly evolving environments1.

For the use case of predicting sales numbers, model accuracy is crucial, because the production model is required to keep up with market changes. Market changes can affect the demand, supply, price, and preference of the products, and thus influence the sales numbers. If the model is not retrained with new data that reflects the market changes, it may become outdated and inaccurate, and fail to capture the patterns and trends of the sales numbers. Therefore, the most likely issue that is causing the steady decline in model accuracy is the lack of model retraining.

The other options are not as likely as option B, because they are not directly related to the model’s ability to adapt to market changes. Option A, poor data quality, may affect the model’s accuracy, but it is not a specific cause of model degradation over time. Option C, too few layers in the model for capturing information, may affect the model’s complexity and expressiveness, but it is not a specific cause of model degradation over time. Option D, incorrect data split ratio during model training, evaluation, validation, and test, may affect the model’s generalization and validation, but it is not a specific cause of model degradation over time. Therefore, option B, lack of model retraining, is the best answer for this question.

References:

Beware Steep Decline: Understanding Model Degradation In Machine Learning Models

Question 71

You are training and deploying updated versions of a regression model with tabular data by using Vertex Al Pipelines. Vertex Al Training Vertex Al Experiments and Vertex Al Endpoints. The model is deployed in a Vertex Al endpoint and your users call the model by using the Vertex Al endpoint. You want to receive an email when the feature data distribution changes significantly, so you can retrigger the training pipeline and deploy an updated version of your model What should you do?

Options:

Use Vertex Al Model Monitoring Enable prediction drift monitoring on the endpoint. and specify a notification email.

In Cloud Logging, create a logs-based alert using the logs in the Vertex Al endpoint. Configure Cloud Logging to send an email when the alert is triggered.

In Cloud Monitoring create a logs-based metric and a threshold alert for the metric. Configure Cloud Monitoring to send an email when the alert is triggered.

Export the container logs of the endpoint to BigQuery Create a Cloud Function to run a SQL query over the exported logs and send an email. Use Cloud Scheduler to trigger the Cloud Function.