Skip to main content

normalize

ferrolearn_preprocess::normalizer

Function normalize 

Source 
pub fn normalize<F: Float>(
    x: &Array2<F>,
    norm: NormType,
    axis: usize,
) -> Result<Array2<F>, FerroError>
Expand description

Scale input vectors individually to unit norm — the standalone, estimator-less API mirroring scikit-learn’s normalize free function (sklearn/preprocessing/_data.py:1866).

With axis == 1 (sklearn’s default) each row (sample) is divided by its norm (L1 = Σ|v|, L2 = √Σv², Max = max|v|); with axis == 0 each column (feature) is normalized instead (sklearn transposes, row-normalizes, and transposes back — :1926-1942, :1971-1972). A row/column whose norm is zero is left unchanged, matching _handle_zeros_in_scale (:1968).

§Errors

Returns FerroError::InvalidParameter if axis is not 0 or 1. Also applies the same check_array input validation as Normalizer’s transform (REQ-2): FerroError::InsufficientSamples for zero rows, and FerroError::InvalidParameter for zero features or any non-finite value (_data.py:1933-1940).