Google Professional Machine Learning Engineer
Get started today
Ultimate access to all questions.
Ultimate access to all questions.
You are tasked with developing a predictive model for detecting potential failures in a critical machine part. Your dataset includes a multivariate time series and corresponding labels that indicate whether a part failure occurred. Recently, you have been exploring various preprocessing and modeling techniques using a Vertex AI Workbench notebook. To effectively monitor your experiments, you need to log data and record artifacts from each run. What steps should you take to properly set up and track your experiments?
Explanation:
The correct answer is option D. This combination provides a comprehensive setup for tracking experiments effectively. Creating a Vertex AI TensorBoard instance allows you to visualize and track metrics, while the Vertex AI SDK sets up and manages experiments. The log_time_series_metrics function is used to track the preprocessed data over time, and the log_metrics function is used to log the loss values, ensuring comprehensive monitoring of both the data and the model's performance. The given references and discussions reinforce that using TensorBoard for these tasks is the recommended practice.