Databricks-ML-Pro Model Development Practice Question
When utilizing Hyperopt with MLflow on Databricks for distributed hyperparameter tuning, which TWO components are strictly required to configure the optimization run properly? (Select TWO)
⚠ Common exam trap
Candidates often include 'an MLflow experiment' or 'a GPU cluster' as required components, while the core logical requirements for Hyperopt are strictly the objective function and the search space.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
An objective function that accepts hyperparameter values and returns a loss metric
Hyperopt optimization routines require an objective function that returns a loss value to minimize and a search space definition that dictates the boundaries of the hyperparameters being explored. Mastering these components is essential for conducting scalable, automated machine learning experiments on Databricks clusters.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
An objective function that accepts hyperparameter values and returns a loss metric
Why this is correct
The objective function is mandatory: Hyperopt's fmin minimises its returned loss value, which must be derived from the hyperparameters passed in. This drives every trial and determines the best configuration, so it is strictly required alongside the search space.
- ✗
A SparkSession configuration object initialized with custom shuffle partitions
Why it's wrong here
Hyperopt needs an objective function and search space; shuffle partition tuning is an unrelated Spark performance setting and is not a required configuration input. It is tempting because adjusting shuffle partitions is a legitimate optimisation for distributed workloads, correctly applied when tuning Spark job performance rather than the search itself.
- ✓
A search space definition specifying the hyperparameters and their distributions
Why this is correct
Hyperopt requires an explicit search space defining each hyperparameter and its distribution (for example uniform, loguniform or choice) so the optimiser knows which values to sample. Without this specification, fmin cannot explore the space, making it strictly required for distributed tuning runs.
- ✗
A pre-registered model URI pointing to an existing MLflow Model Registry entry
Why it's wrong here
Hyperopt requires an objective function and a search space; a Model Registry URI is not needed, since tuning produces new runs rather than consuming a registered model. It is tempting because registry URIs are genuinely required when serving or deploying a tuned model downstream, just not during the tuning run itself.
- ✗
A delta table containing the engineered features for the validation split
Why it's wrong here
A Delta feature table supplies training data but is not a Hyperopt configuration requirement; the objective function and search space are. It is tempting because feature tables are genuinely required in the wider ML pipeline, correctly used when assembling training datasets rather than configuring the optimisation run.
About these practice questions
Courseiva writes every Databricks-ML-Pro question from scratch — 300 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Databricks exam blueprint
This Databricks-ML-Pro practice question is part of Courseiva's free Databricks certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the Databricks-ML-Pro exam.