Module sampling 

Sampling strategies for initial experimental design in Bayesian optimization.

This module provides various sampling methods for generating initial points in the search space before the surrogate model takes over. Proper space-filling designs are critical for Bayesian optimization performance.

Available Methods

• Latin Hypercube Sampling (LHS): Stratified sampling with maximin optimization
• Sobol sequences: Quasi-random low-discrepancy sequences
• Halton sequences: Multi-dimensional quasi-random sequences using prime bases
• Random sampling: Uniform random baseline

Structs

SamplingConfig

Configuration for sampling methods.

Enums

SamplingStrategy

Strategy for generating initial sample points.

Functions

generate_samples

Generate sample points within given bounds.