Skip to main content

SimdArrayOps

numrs2::simd_optimize::simd_traits

Trait SimdArrayOps 

Source 
pub trait SimdArrayOps {
    // Required methods
    fn simd_add(&self, other: &Self) -> Result<Array<f32>>;
    fn simd_mul(&self, other: &Self) -> Result<Array<f32>>;
    fn simd_sum(&self) -> f32;
    fn simd_exp(&self) -> Array<f32>;
    fn simd_log(&self) -> Array<f32>;
    fn simd_sin_cos(&self) -> (Array<f32>, Array<f32>);
    fn simd_matmul(&self, other: &Self) -> Result<Array<f32>>;
    fn simd_dot(&self, other: &Self) -> Result<f32>;
    fn simd_copy(&self) -> Result<Array<f32>>;
}
Expand description

SIMD-optimized extension methods for Array<f32>

These methods provide high-performance SIMD implementations that can be used to accelerate numerical computations on x86_64 and ARM architectures.

Required Methods§

Source

fn simd_add(&self, other: &Self) -> Result<Array<f32>>

SIMD-optimized element-wise addition

Source

fn simd_mul(&self, other: &Self) -> Result<Array<f32>>

SIMD-optimized element-wise multiplication

Source

fn simd_sum(&self) -> f32

SIMD-optimized sum reduction

Source

fn simd_exp(&self) -> Array<f32>

SIMD-optimized exponential function

Source

fn simd_log(&self) -> Array<f32>

SIMD-optimized logarithm function

Source

fn simd_sin_cos(&self) -> (Array<f32>, Array<f32>)

SIMD-optimized trigonometric functions

Source

fn simd_matmul(&self, other: &Self) -> Result<Array<f32>>

SIMD-optimized matrix multiplication

Source

fn simd_dot(&self, other: &Self) -> Result<f32>

SIMD-optimized dot product

Source

fn simd_copy(&self) -> Result<Array<f32>>

SIMD-optimized memory copy

Dyn Compatibility§

This trait is not dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety".

Implementors§