Question

A data scientist is building a Spark ML pipeline for binary classification using LogisticRegression. They want to tune the model over both the regParam and elasticNetParam hyperparameters using CrossValidator. Which object must they construct to define the set of hyperparameter combinations the CrossValidator will search over?

Accepted Answer

A ParamGridBuilder, adding a grid for lr.regParam and lr.elasticNetParam, then calling .build()

Answer

A Python dictionary mapping parameter names to lists of values, passed directly to CrossValidator

Answer

A Hyperopt search space using hp.uniform for each parameter

Answer

A VectorAssembler configured with the two hyperparameters as input columns