ferray-random

Random number generation and distributions for the ferray scientific computing library.

What's in this crate

• Generator API: Generator type with pluggable BitGenerator backends
• BitGenerators: Xoshiro256**, PCG64, Philox, MT19937, PCG64DXSM, SFC64
• SeedSequence: explicit NumPy-equivalent type for reproducible parallel seeding
• 30+ distributions: Normal, Uniform, Exponential, Poisson, Binomial, Gamma, Beta, Chi-squared, Student-t (student_t / standard_t), Laplace, Weibull, Zipf, noncentral_chisquare, noncentral_f, and more
• Permutations: shuffle, permutation, choice (with/without replacement, weighted)
• Parallel generation: standard_normal_parallel with Rayon + jump-ahead
• Deterministic: All output is reproducible given the same seed

Usage

use ferray_random::{default_rng_seeded, Generator};

let mut rng = default_rng_seeded(42);

// Uniform [0, 1)
let samples = rng.random(1000).unwrap();

// Standard normal
let normals = rng.standard_normal(1000).unwrap();

// Integers in [0, 10)
let ints = rng.integers(0, 10, 100).unwrap();

This crate is re-exported through the main ferray crate with the random feature.

License

MIT OR Apache-2.0