Vertex AI & Pipelines Quiz - Competitive Exams GCP_ML_Engineer

Free practice quiz: Vertex AI & Pipelines (Competitive Exams GCP_ML_Engineer). 57 multiple-choice questions with answers and explanations. Play online, no sign-up needed.

This Competitive Exams GCP_ML_Engineer practice quiz has 57 multiple-choice questions on Vertex AI & Pipelines. Try the 5 sample questions below (tap an option to check yourself), then play the full quiz online. No sign-up needed.

Sample questions

1. In Vertex AI Pipelines, what is the primary purpose of a PipelineJob?

   1. To monitor model drift in production
   2. To define and execute a multi-step ML workflow as a directed acyclic graph
   3. To serve a trained model as an online prediction endpoint
   4. To tune hyperparameters automatically
   Show answer

   Answer: B. To define and execute a multi-step ML workflow as a directed acyclic graph

2. Which Vertex AI feature allows you to run a pipeline on a schedule without external orchestration tools?

   1. Vertex AI Endpoint schedules
   2. Vertex AI Pipelines schedules
   3. Vertex AI Model Registry schedules
   4. Vertex AI Workbench schedules
   Show answer

   Answer: B. Vertex AI Pipelines schedules

3. Which of the following best describes the relationship between Vertex AI Pipelines and Kubeflow Pipelines?

   1. Vertex AI Pipelines is a managed service that can run pipelines authored with the Kubeflow Pipelines SDK
   2. Kubeflow Pipelines is a proprietary Google product
   3. Vertex AI Pipelines is a fork of Kubeflow Pipelines that runs only on-premises
   4. They are unrelated and cannot be used together
   Show answer

   Answer: A. Vertex AI Pipelines is a managed service that can run pipelines authored with the Kubeflow Pipelines SDK

4. Which of the following is a key benefit of using Vertex AI Pipelines for MLOps?

   1. It eliminates the need for any containerization of machine learning code.
   2. It provides automated lineage tracking and reproducibility by recording artifacts, parameters, and executions.
   3. It allows pipelines to run only on a single VM, simplifying resource management.
   4. It automatically selects the best machine learning algorithm for a given dataset.
   Show answer

   Answer: B. It provides automated lineage tracking and reproducibility by recording artifacts, parameters, and executions.

5. Which Vertex AI feature allows you to compare two pipeline runs side-by-side, including parameter values and metrics, to identify regressions?

   1. Vertex ML Metadata
   2. Vertex AI Model Registry
   3. Vertex AI Feature Store
   4. Vertex AI Experiments
   Show answer

   Answer: D. Vertex AI Experiments

More questions in this quiz

Answers and explanations are in the full quiz.

1. Which Vertex AI component allows you to run a custom container without managing the underlying infrastructure and is commonly used as a pipeline step?
2. Which of the following is NOT a valid way to trigger a Vertex AI Pipeline run?
3. What does the Vertex AI Pipelines compiler translate pipeline definitions into before execution?
4. Which Vertex AI service provides a central repository to track model versions, metadata, and lineage across pipeline runs?
5. In a Vertex AI Pipeline, what is the role of a pipeline parameter?
6. Which caching behavior occurs by default in Vertex AI Pipelines when a step is re-executed with identical inputs and code?
7. What is the purpose of the Vertex AI Pipelines 'execution' resource?
8. Which of the following best describes a Vertex AI 'component' in the context of pipelines?
9. When using Vertex AI Pipelines with Kubeflow Pipelines SDK, which decorator is used to define a component from a Python function?
10. What is the purpose of Vertex AI Metadata in the context of pipelines?
11. In Vertex AI Pipelines, what does the term 'artifact' refer to?
12. Which component of Vertex AI is used to automatically tune hyperparameters as part of a pipeline?
13. What is a key advantage of using Vertex AI Pipelines over manually running training scripts?
14. In a Vertex AI Pipeline, what is the function of the root task or 'root' execution?
15. Which API is used to submit a pipeline run programmatically in Vertex AI?
16. What is the recommended way to pass large datasets between pipeline steps in Vertex AI Pipelines?
17. Which of the following statements about Vertex AI Pipelines and custom jobs is correct?
18. What is the purpose of the 'enable_caching' parameter when submitting a Vertex AI Pipeline run?
19. Which Vertex AI resource is used to store and manage the container images for pipeline components?
20. In Vertex AI Pipelines, what is the role of a service account?
21. In Vertex AI Pipelines, what is the primary role of the PipelineJob resource?
22. What is the effect of setting the pipeline root to a Cloud Storage path when submitting a Vertex AI Pipeline run?
23. In the context of Vertex AI Pipelines, what does an 'execution' resource represent?
24. Which metadata store concept in Vertex AI represents a specific run of a pipeline or a step, distinct from the reusable definition?
25. In a Vertex AI Pipeline, what is the role of the 'dsl.Condition' construct?
26. What is the primary function of the 'dsl.ExitHandler' in a Vertex AI Pipeline?
27. Which of the following best describes a 'task' in a Vertex AI Pipeline?
28. When defining a pipeline with the Kubeflow Pipelines SDK v2, which function is used to instantiate a pipeline and produce a compiled YAML?
29. In Vertex AI Pipelines, what is the purpose of the 'output' parameter in a component specification?
30. What is the purpose of the 'metadata' parameter when defining a component in the Kubeflow Pipelines SDK?
31. In Vertex AI Pipelines, what is the role of a 'task' as opposed to a 'component'?
32. What is the purpose of the 'kfp.dsl.ParallelFor' construct in a Vertex AI Pipeline?
33. In the context of Vertex AI Pipelines, what is a 'component'?
34. Which of the following statements about pipeline parameters in Vertex AI Pipelines is correct?
35. What is the purpose of the 'execution' resource in Vertex AI Metadata?
36. Which of the following is a valid way to pass data between steps in a Vertex AI Pipeline?
37. What is the role of the 'tfx' (TensorFlow Extended) integration in Vertex AI Pipelines?
38. In Vertex AI Pipelines, what is the purpose of the 'enable_caching' parameter when submitting a pipeline run?
39. What is the primary function of the Vertex AI Pipelines 'compiler'?
40. Which of the following is a valid trigger for a Vertex AI Pipeline run?
41. In Vertex AI Pipelines, what is the purpose of the 'service account' specified when submitting a pipeline job?
42. When defining a pipeline with KFP SDK v2, what is the correct way to declare a parameter that has a default value and a type of float?
43. In Vertex AI Pipelines, what is the purpose of the `dsl.If` construct?
44. Which Vertex AI Pipelines feature allows you to resume a failed pipeline run from the last successful step rather than restarting from the beginning?
45. When submitting a Vertex AI Pipeline run, what does the `pipeline_root` parameter specify?
46. Which of the following best describes the relationship between a Vertex AI Pipeline *task* and an *execution* in Vertex ML Metadata?
47. In Vertex AI Pipelines, what is the recommended way to pass a small string value that should be configurable at runtime without modifying the pipeline definition?
48. Which Vertex AI service can be used to track pipeline runs and compare metrics across different pipeline executions?
49. When defining a custom component in KFP v2, which base image is commonly used for lightweight Python components?
50. Which of the following is a valid way to specify the machine type for a pipeline component that runs a custom training job?
51. In Vertex AI Pipelines, which of the following is a valid trigger for a pipeline run?
52. In Vertex AI Pipelines, what is the recommended way to pass large datasets between pipeline steps?

More quizzes

• Data Engineering for ML▶ Play
• ML Problem Framing▶ Play
• Model Deployment & Monitoring▶ Play

Browse all GCP_ML_Engineer quizzes

More to play