How to log time-series data and track artifacts in Vertex AI?

You are developing a model to predict whether a failure will occur in a critical machine part. You have a dataset consisting of a multivariate time series and labels indicating whether the machine part failed. You recently started experimenting with a few different preprocessing and modeling approaches in a Vertex AI Workbench notebook. You want to log data and track artifacts from each run. How should you set up your experiments?

1. Use the Vertex AI SDK to create an experiment and set up Vertex ML Metadata.
1. Use the Vertex AI SDK to create an experiment and set up Vertex ML Metadata.
1. Create a Vertex AI TensorBoard instance and use the Vertex AI SDK to create an experiment and associate the TensorBoard instance. Source Reference Answer
1. Create a Vertex AI TensorBoard instance, and use the Vertex AI SDK to create an experiment and associate the TensorBoard instance.

Community Votes

50%

50% of anonymous learners picked answer C. Votes are pick records left by other test-takers — they are not the verified answer.

Community Insight

The question tests the ability to distinguish between general artifact tracking (ML Metadata) and the specific requirement for logging time-series metrics, which necessitates associating a TensorBoard instance with the experiment.

To log data and track artifacts for a multivariate time series model in Vertex AI, you should create a Vertex AI TensorBoard instance and associate it with your experiment using the Vertex AI SDK. This setup enables the use of log_time_series_metrics for visualization and assign_input_artifacts for lineage tracking.

Selecting the option to only set up Vertex ML Metadata (Option B) is incorrect because, while it tracks lineage, it does not support the logging and visualization of time-series metrics required for the multivariate dataset.

Community Discussion (8 comments)

gscharly 👍 6 Selected: C

log_time_series_metrics requires setting Tensorboard: https://cloud.google.com/vertex-ai/docs/experiments/log-data assign_input_artifacts can be used to track input data: https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/experiments/get_started_with_vertex_experiments.ipynb

Dirtie_Sinkie 👍 1 Selected: C

C sounds more correct

tungdeptraiqua 👍 3 Selected: B

A and B are the same

fitri001 👍 4 Selected: B

Vertex AI Experiment and ML Metadata: This is the foundation for tracking experiments and artifacts within Vertex AI.expand_more Creating an experiment allows you to group related runs and log data associated with those runs. ML Metadata helps manage the lineage of data and models used in your experiments.expand_more Logging Data: log_time_series_metrics: This function is specifically designed for tracking time-series data, making it suitable for logging the preprocessed multivariate time series data in your experiment. log_metrics: This function is appropriate for logging loss values during model training. It can handle numerical values like loss efficiently. By combining these techniques, you can effectively track both the preprocessed data (time series) and the training performance metrics (loss values) within your Vertex AI Experiment.

omermahgoub 👍 2 Selected: B

Why B? 1. Experiment Creation: Vertex AI SDK establishes a context for grouping your training runs and facilitates experiment management. 2. By setting up Vertex ML Metadata (only can be done when creating an experiment with the Vertex AI SDK), you enable tracking of artifacts and metrics associated with each experiment run. 3. log_time_series_metrics function is well-suited for tracking the preprocessed multivariate time series data associated with each experiment run. This allows you to analyze how preprocessing impacts model performance.

Yan_X 👍 2 Selected: B

B The assign_input_artifacts method is used to associate input artifacts with an experiment, that is not used for log time series and labels. A and B is just with a minor typo (metric vs merric), so select B.

guilhermebutzke 👍 2 Selected: C

My Answer: C assign_input_artifact method is a method to Vertex Ai Experiment to track the preprocessed data while log_time_series_metrics is a function of Vertex AI TensorBoard to log metrics along time. look: https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/experiments/build_model_experimentation_lineage_with_prebuild_code.ipynb https://github.com/GoogleCloudPlatform/vertex-ai-samples/blob/main/notebooks/official/experiments/comparing_local_trained_models.ipynb

b1a8fae 👍 2 Selected: C

C. Tensorboard for experimentation and comparison of different model runs. assign_input_artifacts to track preprocessed data, since it links artifacts as inputs to the execution. https://cloud.google.com/python/docs/reference/aiplatform/latest/google.cloud.aiplatform.Execution#google_cloud_aiplatform_Execution_assign_input_artifacts Using log_time_series_metrics would make sense if what we were doing is logging a metric, which we aren't when we track the preprocessed data not yet ran by the model.

Comments & Corrections

No comments yet — spotted an error or have a note? Share it below.

Expert Analysis

Why the Answer Is Correct

The correct approach involves creating a Vertex AI TensorBoard instance and associating it with the experiment. As noted in community comments, the function log_time_series_metrics is specifically designed for the multivariate time series data mentioned in the scenario and explicitly requires a TensorBoard instance to function. Furthermore, the Vertex AI SDK allows tracking input artifacts via assign_input_artifacts, satisfying the requirement to track artifacts alongside the metric logging.

Why the Other Options Are Wrong

Options A and B focus solely on creating an experiment and setting up Vertex ML Metadata. While ML Metadata is the underlying engine for tracking artifacts and lineage, it does not provide the infrastructure to visualize time-series metrics. Given the specific mention of a "multivariate time series," the solution must support time-series logging, which is a capability provided by TensorBoard, not ML Metadata alone.

Community Comment Notes

There was a 50/50 split in the voting, with some users advocating for B due to the importance of ML Metadata in tracking artifacts. However, top-rated comments supporting C highlighted the technical dependency of log_time_series_metrics on a TensorBoard instance. Commenters also pointed out that assign_input_artifacts handles the artifact tracking within the experiment, making C the more comprehensive solution for the specific data type described.

Official Reference

Exam Strategy

When you see keywords like "time series" or "metrics over time" in a Vertex AI question, immediately look for options that include creating or associating a TensorBoard instance, as standard experiment logging is insufficient for time-series visualization.

Related Analysis

Practice All PMLE Questions

Access 65 questions with complete answers and detailed explanations.

View Full PMLE Practice Test →

← Back to PMLE Study Guide