Databricks Machine Learning Associate Question 464
Select 3You are working on a machine learning project in Databricks and are using MLflow to track your experiments. Which of the following pieces of information can you view in the MLflow UI?
- A
The hyperparameters used for each experiment run
- B
The code version or Git commit hash associated with each run
- C
The number of training epochs completed in each run
- D
The model artifacts generated during a run
- E
The Spark cluster configuration used for the experiment
Show answer and explanation
Correct answers: A, B, D
Explanation
The MLflow UI is designed to display key information related to experiment tracking, such as hyperparameters, code version, and artifacts. However, details like training epochs or Spark cluster configurations are not logged by default and may require custom logging if needed.
- A. Correct.
The MLflow UI displays logged hyperparameters for each experiment run, as they are critical for reproducibility and analysis.
- B. Correct.
The MLflow UI can show the code version or Git commit hash if it is logged during the run, helping track the exact code used.
- C. Incorrect.
The number of training epochs is not logged automatically by MLflow unless explicitly logged, so it may not always appear in the UI.
- D. Correct.
The MLflow UI provides access to logged artifacts, such as model files or evaluation metrics, which are part of the experiment run.
- E. Incorrect.
The Spark cluster configuration is not tracked or displayed in the MLflow UI, as it is not directly relevant to experiment tracking.