Databricks Certified Machine Learning - Associate
Get started today
Ultimate access to all questions.
Comments
Loading comments...
Ultimate access to all questions.
Consider a scenario where you are tasked with optimizing a Spark ML model using Hyperopt. You have a dataset with 1 million records and a limited computational budget. How would you approach the parallelization of the hyperparameter tuning process to maximize the model's accuracy within the given constraints?
A
Use a single-threaded approach and sequentially run the hyperparameter tuning trials, as parallelization is not necessary for Spark ML models.
B
Parallelize the hyperparameter tuning process using Hyperopt's Trials, but limit the number of concurrent trials to match the available computational resources to avoid exceeding the budget.
C
Increase the number of trials exponentially to maximize the exploration of the hyperparameter space, regardless of the computational budget.
D
Focus solely on the initial choice of hyperparameters and avoid any hyperparameter tuning, as the computational budget is too limited for an effective optimization process.