Skip to main content

hermite_function

Function hermite_function 

Source
pub fn hermite_function<P, E, V, const EXACT_FMA: bool>(x: V, n: u32) -> V
where P: Policy, E: FloatElement, V: FloatVector<Element = E> + SpecializedTranscendentalMath<E>,
Expand description

The runtime-degree twin of hermite_function_n.

The same seed, recurrence and final factor, with the degree as a value. a_k and b_k are computed per step rather than folded. Both are a correctly rounded division and square root either way, so the result agrees with the const form to the bit.

Last built: 2026-09-08 21:35:55 UTC