Google Professional Machine Learning Engineer exam dumps

Google Professional Machine Learning Engineer practice question 379 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 379

Select 3Google Cloud Platform

You are tasked with setting up a CI/CD pipeline for a machine learning model training workflow on Google Cloud. The pipeline must automatically retrain the model whenever new data is uploaded to a Cloud Storage bucket and should use a managed service to handle the execution of the training script. Which combination of components and services would best meet these requirements?

  1. A

    Use Cloud Build to orchestrate the CI/CD pipeline and trigger the training job.

  2. B

    Use Cloud Functions to trigger the pipeline when new data is uploaded to the Cloud Storage bucket.

  3. C

    Use Vertex AI Training for running the training job in a managed environment.

  4. D

    Use Compute Engine to manually provision training infrastructure.

  5. E

    Use Pub/Sub to notify the pipeline when new data is uploaded to the Cloud Storage bucket.

Show answer and explanation

Correct answers: A, B, C

Explanation

The combination of Cloud Build, Cloud Functions, and Vertex AI Training provides an efficient and managed solution for automating the CI/CD pipeline for model retraining. Cloud Build handles the orchestration, Cloud Functions triggers the pipeline based on new data uploads, and Vertex AI Training ensures the training job runs in a managed environment. This setup avoids the need for manual infrastructure management and ensures the pipeline is responsive and scalable.

  • A. Correct.

    Correct: Cloud Build is an ideal choice for orchestrating the CI/CD pipeline as it can automate the workflow and integrate with other Google Cloud services.

  • B. Correct.

    Correct: Cloud Functions can be used to trigger the pipeline automatically when new data is uploaded to the Cloud Storage bucket, ensuring the pipeline starts at the right time without manual intervention.

  • C. Correct.

    Correct: Vertex AI Training provides a managed environment for training machine learning models, removing the need to manually provision and manage infrastructure.

  • D. Incorrect.

    Incorrect: Compute Engine requires manual provisioning and management of resources, which is less efficient and not ideal for a managed CI/CD pipeline.

  • E. Incorrect.

    Incorrect: While Pub/Sub can be used for notifications, it is unnecessary in this scenario because Cloud Functions can directly handle the event trigger from Cloud Storage.

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