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 working with a large-scale dataset and a Spark ML model. You want to optimize the model's hyperparameters using Hyperopt, but you are concerned about the computational cost of running a large number of trials. How would you approach this situation to balance the exploration of the hyperparameter space with the available computational resources?
A
Run a fixed number of trials regardless of the computational cost, as the exploration of the hyperparameter space is more important than optimizing the computational resources.
B
Limit the number of trials to a small fixed value to minimize the computational cost, even if it means not exploring the hyperparameter space thoroughly.
C
Use Hyperopt's adaptive algorithms, such as Tree-structured Parzen Estimator (TPE), to intelligently select the next set of hyperparameters to evaluate based on the results of previous trials, allowing for more efficient exploration of the hyperparameter space.
D
Focus on manual hyperparameter tuning instead of using Hyperopt, as it provides more control over the computational cost while sacrificing the exploration of the hyperparameter space.