Databricks Machine Learning Associate Question 462
Select 4A data scientist is using MLflow with Databricks to track their machine learning experiments. They open the MLflow UI to review the details of a specific run. Which of the following pieces of information can they view directly within the MLflow UI?
- A
The parameters used in the experiment's model training
- B
The hyperparameter tuning grid used for the experiment
- C
The metrics logged during the experiment run
- D
The versions of artifacts associated with the experiment
- E
The source code or notebook used for the experiment
- F
The comparison of metrics between multiple experiment runs
Show answer and explanation
Correct answers: A, C, D, E
Explanation
The MLflow UI provides essential information about individual experiment runs, including parameters, metrics, artifacts, and the source code used for the experiment. However, it does not display comprehensive hyperparameter tuning grids or directly compare metrics across runs unless you switch to the comparison view.
- A. Correct.
Correct. The MLflow UI displays parameters logged during an experiment, such as the values of hyperparameters or other inputs to the model.
- B. Incorrect.
Incorrect. While hyperparameter tuning grids might be part of a larger experiment, they are not directly viewable in the MLflow UI. Only the logged parameters are displayed.
- C. Correct.
Correct. The MLflow UI provides metrics logged during the experiment, such as accuracy, loss, or other performance metrics.
- D. Correct.
Correct. The MLflow UI shows versions of artifacts, such as trained model files or datasets, that were logged during a run.
- E. Correct.
Correct. If the experiment source code or notebook path was logged, it will be visible in the MLflow UI under the 'Source' section.
- F. Incorrect.
Incorrect. While you can compare metrics across runs, this functionality requires using the comparison feature specifically rather than being part of the information displayed for a single run.