pp_snv — Standard Normal Variate (SNV)

Group: Preprocessing · Binding: n4m.sklearn.SNV · C ABI: n4m_pp_snv_*

Description

Standard Normal Variate normalisation.

Parameters

Name	Type	Default
with_mean	bool	True
with_std	bool	True
ddof	int	0

Explanations

Bibliographic source

Barnes, R. J., Dhanoa, M. S. & Lister, S. J. (1989). Standard Normal Variate Transformation and De-trending of Near-Infrared Diffuse Reflectance Spectra. Applied Spectroscopy 43(5), 772–777.

Mathematical principle

Each spectrum \(\mathbf{x}_i\) is centred and scaled by its own row statistics: \(\mathrm{SNV}(\mathbf{x}_i) = (\mathbf{x}_i - \bar{x}_i)/s_i\), where \(\bar{x}_i\) and \(s_i\) are the mean and standard deviation across the wavelengths of that single spectrum. This removes multiplicative scatter and additive baseline shifts on a per-sample basis without needing a reference spectrum, which is why SNV is robust to sample-to-sample path-length variation in diffuse-reflectance NIR.

Implementation

C ABI n4m_pp_snv_* in libn4m (create / apply / destroy lifecycle), wrapped by n4m.sklearn.SNV. The same numerical kernel backs every language binding.

Usage

from n4m.sklearn import SNV
op = SNV()
X_transformed = op.fit_transform(X)

Benchmarks

Adaptive wall-clock per cell measured against full_matrix.csv. Only backends that implement this method are listed; libraries without the method are omitted.

Verdict  ·  ✓ ref / ≈ ref / ~ shape mark a reference-gate pass at strict / relaxed / qualitative tolerance  ·  ✓ bind = pls4all binding agrees with the C++ baseline  ·  ⇄ cross-check = documented by-design selector/RNG/model, noncanonical API/facade convention, or secondary oracle  ·  ✗ divergent  ·  ⚠ error  ·  — not run. The fastest backend per column is marked 🏆.

Reference gate: strict — numeric equivalence (rmse_rel_tol ≤ 1e-12).

1 thread

Backend	Parity	50×250 (ms)	250×50 (ms)
C++ native · libn4m
pls4all.cpp.blas	✓ ref	—	0.04 ms
pls4all.cpp.blas+omp	✓ ref	0.02 ms	0.01 ms
pls4all.cpp.omp	✓ ref	—	0.01 ms🏆
pls4all.cpp.ref	✓ ref	0.02 ms🏆	0.04 ms
Python · pls4all
pls4all.python	✓ bind	0.02 ms	0.02 ms
pls4all.sklearn	✓ bind	0.02 ms	0.02 ms
Python · external
nirs4all	source	0.10 ms	0.12 ms
ref.python_numpy	source	—	—

---

See also: methods index · interactive dashboard