Databricks-ML-Pro Model Development Practice Test 5 — 15 Questions

Question 1

A machine learning engineer is preparing to deploy a model to production using Databricks Model Serving. They want to ensure that the model's input schema is correctly enforced and that the serving endpoint can handle missing values gracefully. Which TWO actions should they take when logging the model to the MLflow Model Registry? (Choose two.)

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Provide a model signature that includes input and output schema, specifying the data types and column names.. To enforce input schema and handle missing values in Databricks Model Serving, the model signature must be explicitly defined with appropriate data types. A signature with nullable types (e.g., `double`) allows missing values to pass validation. Additionally, the signature itself enforces the schema, rejecting incompatible inputs. The other options either do not directly address schema enforcement or are not actions taken during logging. The correct actions are to provide a signature and ensure it allows nulls.

Answer

Log the model with an input example that contains representative data, including missing values, to infer the signature.

Answer

Include a `requirements.txt` file with the model to specify dependencies, ensuring the serving environment can load the model.

Answer

Set the `mlflow.pyfunc` model's `predict` method to handle missing values by imputing them before passing to the underlying model.

Question 2

A machine learning engineer is training a PyTorch model on a Databricks cluster and needs to distribute the training across multiple worker nodes. Which framework should be integrated natively within Databricks to handle this distributed deep learning workflow efficiently?

Accepted Answer

TorchDistributor. TorchDistributor is the native Databricks library designed to launch distributed PyTorch training jobs seamlessly across cluster nodes using standard PyTorch native CLI commands and environment configurations. Understanding distributed training orchestration on Databricks is crucial for scaling deep learning pipelines on large datasets without manually managing cluster communication sockets.

Answer

Databricks Feature Store Client

Answer

MLflow Model Registry

Answer

Spark MLlib Pipeline

Question 3

An ML engineer is training an XGBoost model on Databricks and wants to leverage hyperparameter tuning using Hyperopt while automatically logging all trial parameters, metrics, and models to MLflow. Which built-in MLflow function should be used to achieve this automatic integration?

Accepted Answer

mlflow.xgboost.autolog(). mlflow.xgboost.autolog() automatically logs parameters, metrics, and trained artifacts during XGBoost training routines without requiring manual logging statements inside the training loop. This function streamlines model development workflows by ensuring complete lineage tracking and experiment reproducibility across extensive hyperparameter tuning sweeps.

Answer

mlflow.spark.autolog()

Answer

mlflow.sklearn.autolog()

Answer

mlflow.register_model()

Question 4

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)

Accepted Answer

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

A SparkSession configuration object initialized with custom shuffle partitions

Answer

A pre-registered model URI pointing to an existing MLflow Model Registry entry

Answer

A delta table containing the engineered features for the validation split

Question 5

Refer to the exhibit. A data scientist is logging their model training process. Which statement accurately describes the storage location of the artifacts referenced in the code snippet?

Accepted Answer

The model artifact created by log_model is stored in the MLflow tracking store's internal directory structure.. MLflow handles artifacts differently based on the log function used. While 'log_artifact' takes a explicit path (often DBFS or local filesystem), 'log_model' packages the model with its metadata and dependencies into a standard MLflow directory structure. Understanding this distinction is essential for production deployments, as model registry and deployment tools rely on the specific internal structure generated by the 'log_model' function, not just raw file paths.

Answer

All items are stored in the same local directory on the driver node.

Answer

The log_artifact command uploads the weights directly to the Model Registry.

Answer

The log_model function is equivalent to copying the file directly to DBFS.

Question 6

When developing a machine learning pipeline on Databricks, which feature provides the most effective way to track the lineage of a model from the raw data used for training to the final deployment?

Accepted Answer

Using MLflow Tracking integrated with Unity Catalog lineage.. MLflow Tracking and the Unity Catalog integration are the cornerstones of model lineage in Databricks. Tracking allows developers to log parameters, code versions, and data snapshots, while Unity Catalog provides governance and data lineage. Together, these tools ensure full traceability, which is a mandatory requirement for compliance and auditing in enterprise-grade machine learning systems where understanding how a model reached its current state is critical.

Answer

Manually maintaining a spreadsheet documenting training data versions.

Answer

Storing training data in a local folder on the driver node.

Answer

Creating a new database for every model training run.

Question 7

A machine learning team is transitioning from local notebooks to Databricks. They want to ensure their code is modular and reusable. Which THREE practices should they implement?

Accepted Answer

Package common utility code into Python wheels.. Transitioning to Databricks requires moving away from monolithic notebooks toward modular code structures. Using Delta Lake for data consistency, moving logic into Python wheels, and utilizing Repos for version control are the standard industry practices. These steps ensure that code is maintainable, testable, and capable of being integrated into automated CI/CD pipelines, which is the hallmark of mature MLOps practices within a Databricks workspace.

Answer

Keep all training, preprocessing, and evaluation code in a single notebook.

Answer

Hardcode all file paths and credentials in every notebook.

Question 8

Which of the following describes the correct usage of the MLflow 'log_param' function in a Databricks environment?

Accepted Answer

It logs individual hyperparameter values, such as learning rates or tree depth, to the active run.. Logging parameters is a best practice for experiment reproducibility. By calling 'log_param' for every hyperparameter, you ensure that every run is documented with the exact settings used. This allows for rigorous comparison between runs. In a production environment, this metadata is what enables the team to identify which model version is associated with specific hyperparameter configurations, facilitating model auditing and governance.

Answer

It should be called after the model is trained to log the final weights.

Answer

It automatically logs the entire training dataset for lineage tracking.

Answer

It is only available for custom models and not for Spark MLlib models.

Question 9

When evaluating a classification model on Databricks, a team needs to generate a custom performance report that is not natively provided by MLflow. What is the recommended strategy to ensure this report is persisted and associated with the training run?

Accepted Answer

Save the report to a local temporary file and use MLflow.log_artifact to upload it to the run.. Logging custom plots and reports as artifacts is the recommended way to enrich MLflow run metadata. By saving visualizations or summary statistics as files (e.g., HTML, PNG, or JSON) and using 'log_artifact', the team ensures that non-standard evaluation metrics are permanently stored. This practice allows stakeholders to review specific model performance indicators without needing to re-run the training code, which is vital for compliance and post-training analysis.

Answer

Print the report to the notebook output and rely on the notebook's command history.

Answer

Store the report in an external database and manually link it via a comment in the run.

Answer

Re-generate the report every time the model is loaded for inference.

Question 10

Which THREE of the following are considered best practices for handling data preprocessing in a Databricks ML pipeline to prevent data leakage?

Accepted Answer

Fit your scaler/transformer exclusively on the training portion of the data.. Preventing data leakage is essential for valid model performance. The key is to calculate statistics (like means or scaling factors) strictly on the training set and apply them to the validation/test sets. Using Spark ML transformers encapsulates this logic, ensuring that the transformation process is consistent and prevents the model from 'seeing' information from the test set during the training phase, leading to accurate performance estimation.

Answer

Calculate global mean imputation values using the entire dataset before splitting.

Answer

Perform feature selection using the entire dataset to maximize model power.

Question 11

Refer to the exhibit. What happens to these logged metrics in MLflow when the training run completes?

Accepted Answer

MLflow keeps all values, enabling the visualization of metrics over time.. MLflow is designed to track metrics over time, which is crucial for monitoring model convergence during training. By logging the same metric key multiple times, MLflow stores each value as a point in a time series. This allows developers to visualize the training progress and identify when the model stops improving, which is critical for implementing early stopping and optimizing training efficiency in deep learning.

Answer

MLflow overwrites the previous value with the latest value.

Answer

The run will throw an error because the metric key is not unique.

Answer

MLflow averages the values and stores the mean.

Question 12

Which Databricks feature is specifically designed to manage the lifecycle of a machine learning model, including versioning, stage transitions, and deployment tracking?

Accepted Answer

MLflow Model Registry.. The MLflow Model Registry is the definitive tool in Databricks for managing the model lifecycle. It allows teams to register models, track versions, and manage stage transitions (e.g., Staging to Production). By providing a centralized, audit-trailed repository, it ensures that only validated models are deployed into production, fulfilling the core requirements of MLOps for governance, reliability, and automated deployment pipelines.

Answer

Unity Catalog.

Answer

Databricks Delta Lake.

Answer

Databricks Repos.

Question 13

A team is building an automated retraining pipeline. They need to ensure that only models exceeding a certain performance threshold are registered. What is the most effective way to implement this logic?

Accepted Answer

Write a Python script that evaluates the model and conditionally calls the Model Registry API.. Integrating conditional logic into the retraining script using the MLflow API allows for programmatic model governance. By evaluating the model against validation data and only calling 'register_model' if the performance exceeds the threshold, the team prevents poor-quality models from entering the registry. This automated gatekeeping is vital for maintaining the health of the production pipeline and ensuring that only high-performing models proceed to deployment.

Answer

Manually inspect all models in the MLflow UI before registering them.

Answer

Register every trained model and let the production system filter them.

Answer

Use MLflow's 'auto-register' feature that registers all models by default.

Question 14

A machine learning engineer is training a scikit-learn model on Databricks and wants to automatically log hyperparameters, metrics, and the trained artifact without writing extensive boilerplate logging code. Which approach should the engineer use?

Accepted Answer

Execute mlflow.sklearn.autolog() prior to initiating the model training fit process to automatically capture all runs.. Autologging automatically captures all relevant parameters, metrics, and models without requiring manual mlflow.log_param calls. This ensures consistency, prevents human error during tracking, and natively integrates with Spark dataframes. It significantly accelerates the model development lifecycle on Databricks by standardizing the capture of experimentation metadata across distributed runs.

Answer

Manually invoke mlflow.start_run() and explicitly call mlflow.log_metric() for every single training epoch iteration.

Answer

Store all model metrics directly inside a temporary Delta table and sync them to MLflow during the final evaluation phase.

Answer

Configure the cluster environment variables to automatically intercept stdout and parse training metrics using regex.

Question 15

Refer to the exhibit. A data scientist is preparing to log a model. What is the primary benefit of including the explicit 'signature' provided in the exhibit during the mlflow.log_model process?

Accepted Answer

It allows the serving endpoint to perform automatic data type validation on incoming requests.. The model signature acts as a contract between the model and its environment. By explicitly defining the input and output schema, Databricks can perform validation during inference to prevent runtime errors caused by mismatched data types. This is critical for production pipelines where data quality might fluctuate, ensuring that the model consumes and produces the expected data formats reliably.

Answer

It enables automatic hyperparameter tuning during the next training iteration.

Answer

It automatically encrypts the model weights for secure storage in the registry.

Answer

It forces the model to be converted into a serialized format like ONNX.