Free Google Professional-Machine-Learning-Engineer Actual Exam Questions - Question 12 Discussion

B. Cloud Composer can automate the whole training pipeline, including multiple runs, but it doesn’t inherently provide a dashboard for comparing metrics. You’d still need to integrate with another tool for visualization. That makes it less straightforward for the question’s goal of having one place to compare models.

C is tempting since AI Platform supports multiple jobs, but managing and comparing many jobs manually can get messy without a proper experiment tracking setup, which Kubeflow offers natively.

It’s A because AutoML Tables lets you easily try different models and provides a dashboard to compare metrics all in one place, which fits the need for exploring multiple architectures and tracking results.

C imo, running multiple training jobs with similar names on AI Platform lets you track them in the GCP console. It’s simple and integrates well without needing extra setup like Kubeflow.

D (Cloud Composer doesn’t offer built-in metric comparison dashboards)

Makes sense to pick D here, as Kubeflow Pipelines is specifically built for organizing multiple runs and comparing metrics in one place, which fits the dashboard need. D

D—Kubeflow Pipelines is made for organizing and comparing model runs easily.

Probably B too. Cloud Composer automates workflows, so you can schedule and run multiple training jobs automatically, making it easier to keep track of datasets and model versions. While it’s not exactly a built-in dashboard, it integrates well with other GCP monitoring tools to compare evaluation metrics.

It’s different from D because Kubeflow Pipelines focuses more on experiment tracking within the pipeline itself, but if you want solid orchestration and data management across runs, Composer could be a strong choice as well.

D seems right since Kubeflow Pipelines is built for managing and comparing experiments easily. D

B could work since Cloud Composer automates workflows and can link data storage with training runs, making it easier to compare results later. But does it really provide a built-in dashboard for metrics comparison like Kubeflow Pipelines?

Option D seems best since Kubeflow Pipelines can manage multiple runs neatly and integrate with tools to track metrics, plus it handles data versioning better than just running jobs separately.

Maybe B makes sense too since Cloud Composer can automate workflows, including running multiple trainings and managing data steps, which helps compare results in one place. It’s more about orchestration than just model runs.

C imo, running multiple training jobs on AI Platform with similar job names can help keep things organized and makes it easier to compare evaluation metrics since you can pull results from jobs with consistent naming conventions. It’s not as automated as Cloud Composer or Kubeflow but straightforward for managing multiple architectures.

Plus, storing training data separately in Cloud Storage works well regardless of which option you pick, so you can still keep your data safe and accessible while handling the modeling side on AI Platform.

It’s B because Cloud Composer can automate multiple training runs and orchestrate workflows, making it easier to manage experiments and track metrics in one place without manual job naming.

C/D? I think C could work if you just want to run multiple jobs and name them similarly to track, but that’s pretty manual and messy for comparing metrics in one dashboard. Kubeflow (D) is built to handle organizing runs, comparing metrics clearly, and you can integrate storage separately for data. B automates the runs but doesn’t really give you a unified dashboard for metrics out of the box. So if you want a clean comparison and organization, D seems like the better fit overall.

D, seems like Kubeflow Pipelines fits best for organizing multiple runs and comparing metrics.