Module random 

Random number generation, shuffling, and weighted sampling.

Provides seeded RNG construction, Fisher-Yates shuffle, and weighted random sampling utilities.

Reproducibility

For reproducible experiments, use create_rng with a fixed seed. The underlying algorithm (SmallRng) is deterministic for a given seed on the same platform.

Structs

WeightedSampler

Pre-computed weighted sampler for O(log n) repeated sampling.

Functions

create_rng

Creates a fast, seeded random number generator.

shuffle

Fisher-Yates (Durstenfeld) in-place shuffle.

shuffled_indices

Returns a shuffled index permutation of [0, n).

weighted_choose

Selects a random index weighted by the given weights.