A data scientist set up a machine learning pipeline to automatically log a data visualization with each run. They now want to view the visualizations in Databricks. Which of the following locations…

Question

Options

• AThe MLflow Model Registry Model paqe
• BThe Artifacts section of the MLflow Experiment page
• CLogged data visualizations cannot be viewed in Databricks
• DThe Artifacts section of the MLflow Run page
• EThe Figures section of the MLflow Run page

• A
  3% (1)
• B
  6% (2)
• D
  13% (4)
• E
  78% (25)

Explanation

Option E is correct because MLflow provides a dedicated Figures section on the Run page specifically for visualizations logged via mlflow.log_figure(). Databricks surfaces this as its own tab, separate from generic artifacts, making it the proper landing spot for programmatically logged plots.

Why the distractors are wrong:

• A (Model Registry) is for managing versioned, registered models - not run-level artifacts or visuals.
• B (Artifacts section of the Experiment page) is incorrect because artifacts are scoped to individual runs, not the experiment level; experiments aggregate runs but don't have their own Artifacts tab.
• C is simply false - Databricks absolutely supports viewing logged visualizations.
• D (Artifacts section of the Run page) is the close distractor: generic files logged with mlflow.log_artifact() do appear there, but figures logged with mlflow.log_figure() get their own Figures tab, not the Artifacts tab.

Memory tip: Think "figures go to Figures" - mlflow.log_figure() → Figures tab; mlflow.log_artifact() → Artifacts tab. The dedicated tab exists precisely to render visuals inline rather than forcing you to download a file.

No community discussion yet for this question.