Databricks Machine Learning Associate Question 58
Select 3A data scientist is tasked with building a predictive model for customer churn using Databricks AutoML. After running AutoML on their dataset, they notice that several models were automatically trained and evaluated. Which of the following steps are performed automatically by Databricks AutoML during this process?
- A
Splitting the dataset into training, validation, and test sets
- B
Hyperparameter tuning for all trained models
- C
Feature engineering, such as one-hot encoding for categorical features
- D
Generating Python notebooks to allow users to review the model training process
- E
Deploying the best-performing model directly into production
Show answer and explanation
Correct answers: A, C, D
Explanation
Databricks AutoML simplifies the machine learning workflow by automating tasks like dataset splitting, feature engineering, and model training. It also provides transparency through the generation of Python notebooks. However, tasks like exhaustive hyperparameter tuning and model deployment need to be handled separately by users.
- A. Correct.
Correct: Databricks AutoML automatically splits the input dataset into training, validation, and test sets as part of its workflow.
- B. Incorrect.
Incorrect: Databricks AutoML does not perform exhaustive hyperparameter tuning for all models. Instead, it uses a basic tuning strategy to quickly evaluate models.
- C. Correct.
Correct: Databricks AutoML performs basic feature engineering, including transformations like one-hot encoding for categorical features.
- D. Correct.
Correct: Databricks AutoML generates Python notebooks that document the model training process, enabling users to understand and modify the results.
- E. Incorrect.
Incorrect: Databricks AutoML does not directly deploy models into production. Deployment requires additional steps performed by the user.