Google Professional Machine Learning Engineer exam dumps

Google Professional Machine Learning Engineer practice question 122 of 522

Professional Machine Learning Engineer. Professional level, Google Cloud. Free question with the correct answer and a full explanation.

Google Professional Machine Learning Engineer Question 122

Select 4Google Cloud Platform

You are building a machine learning pipeline to predict customer churn using Google Cloud. You decide to use the Vertex AI Feature Store to manage your features. You need to create a feature store that consolidates features from multiple datasets, such as transaction history, user profile data, and website interaction logs. Which of the following steps should you take to ensure your feature store is set up correctly and ready for use?

  1. A

    Define a feature store and create entity types for each dataset, such as 'transactions', 'profiles', and 'interactions'.

  2. B

    Ingest features into the feature store by specifying the entity type and corresponding feature values.

  3. C

    Manually engineer all features and store them in a custom Google Cloud Storage bucket.

  4. D

    Use a unique identifier for each entity type to link related features together, such as a customer ID.

  5. E

    Configure a Vertex AI Feature Store online serving endpoint for real-time predictions, if required.

Show answer and explanation

Correct answers: A, B, D, E

Explanation

To set up a Vertex AI Feature Store, you need to define the store and entity types, ingest features, and link related features using unique identifiers. Additionally, configuring an online serving endpoint is necessary for real-time predictions. However, manually storing features in a storage bucket bypasses the benefits provided by Vertex AI Feature Store, such as feature sharing, versioning, and online serving.

  • A. Correct.

    Correct: Defining a feature store and creating entity types ensures you logically organize features for different datasets.

  • B. Correct.

    Correct: Ingesting features into the feature store is a necessary step to make the features available for training and serving.

  • C. Incorrect.

    Incorrect: While manual feature engineering is sometimes required, storing features in Google Cloud Storage does not leverage Vertex AI Feature Store's capabilities like versioning or serving.

  • D. Correct.

    Correct: Using a unique identifier, such as a customer ID, is crucial for linking related features across entity types.

  • E. Correct.

    Correct: Configuring an online serving endpoint in Vertex AI Feature Store is important for real-time inference when needed.

Timed practice exam

Take a Google Professional Machine Learning Engineer practice test under exam conditions

60 questions in 120 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam