Struct ZScoreNormF32 

pub struct ZScoreNormF32 { /* private fields */ }

Z-score normalizer backed by EWMA variance.

Returns (sample - mean) / std_dev using exponentially weighted estimates of mean and standard deviation. Useful for online feature normalization before feeding into adaptive filters or learners.

Examples

use nexus_stats_core::normalization::ZScoreNormF32;

let mut zs = ZScoreNormF32::builder().span(20).build().unwrap();
for i in 0..100 {
    let _ = zs.update(100.0 as f32 + i as f32);
}
let z = zs.update(150.0).unwrap();
assert!(z.is_some());

Implementations

impl ZScoreNormF32

pub fn builder() -> ZScoreNormF32Builder

Creates a builder.

pub fn update(&mut self, sample: f32) -> Result<Option<f32>, DataError>

Feeds a sample. Returns the z-score once primed.

If the standard deviation is zero (constant stream), returns 0.0.

Errors

Returns DataError::NotANumber if the sample is NaN, or DataError::Infinite if the sample is infinite.

pub fn normalize(&self, value: f32) -> Option<f32>

Normalizes an arbitrary value against current statistics without updating state.

Returns None if not primed.

pub fn mean(&self) -> Option<f32>

Current smoothed mean, or None if not primed.

pub fn std_dev(&self) -> Option<f32>

Current exponentially weighted standard deviation, or None if not primed.

pub fn count(&self) -> u64

Number of samples processed.

pub fn is_primed(&self) -> bool

Whether enough samples have been observed.

pub fn reset(&mut self)

Resets accumulated state. Parameters unchanged.

Trait Implementations

impl Clone for ZScoreNormF32

fn clone(&self) -> ZScoreNormF32

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ZScoreNormF32

fn fmt(&self, f: &mut Formatter<'_>) -> Result<(), Error>

Formats the value using the given formatter.