Skip to main content

simd_standardize_f32

scirs2_core::simd::preprocessing

Function simd_standardize_f32 

Source 
pub fn simd_standardize_f32(input: &ArrayView1<'_, f32>) -> Array1<f32>
Expand description

SIMD-accelerated standardization for f32 arrays

Returns (x - mean) / std for each element (z-score normalization). If std is zero or NaN, returns a zero array.

§Arguments

  • input - Input array to standardize

§Returns

Standardized array with mean=0 and std=1 (or zero array if input std is zero)

§Performance

  • f32 (100K elements): 1.22x faster than NumPy ✓
  • f64 (100K elements): 0.88x-0.94x vs NumPy

This implementation uses direct SIMD pointer writes for optimal performance.