argmax_simd

scirs2_core::ndarray_ext::reduction

Function argmax_simd 

Source 
pub fn argmax_simd<F>(x: &ArrayView1<'_, F>) -> Option<usize>
where
    F: Float + SimdUnifiedOps,
Expand description

Find the index of the maximum element in a 1D array with SIMD acceleration.

This function uses SIMD operations when beneficial for performance, providing significant speedups for large arrays on supported platforms (AVX2, NEON).

§Arguments

  • x - Input 1D array

§Returns

  • Some(index) - Index of the maximum element
  • None - If the array is empty

§Performance

  • f32: ~2-3x faster than scalar for arrays > 1000 elements
  • f64: ~2-3x faster than scalar for arrays > 1000 elements
  • Automatically selects SIMD or scalar based on array size and platform capabilities

§Examples

use scirs2_core::ndarray::array;
use scirs2_core::ndarray_ext::reduction::argmax_simd;

let x = array![3.0f32, 1.0, 4.0, 1.0, 5.0, 9.0, 2.0];
let idx = argmax_simd(&x.view()).unwrap();
assert_eq!(idx, 5); // Index of maximum value (9.0)