Scikit-learn–Styled Optimizers

Hyperactive provides sklearn-style search using ParameterGrid / ParameterSampler, with evaluation routed through Hyperactive experiments. This preserves familiar configuration while keeping the v5 architecture’s separation between search (optimizer) and evaluation (experiment).

Key characteristics:

• Sklearn-style search spaces via ParameterGrid / ParameterSampler
• Evaluation via a Hyperactive Experiment (typically SklearnCvExperiment)
• Parallelism via optimizer backends (backend, backend_params)
• Results exposed on the optimizer (best_params_, best_score_, etc.)