pp_savgol — Savitzky–Golay smoothing / derivative

Group: Preprocessing · Binding: n4m.sklearn.SavitzkyGolay · C ABI: n4m_pp_savgol_*

Description

scipy.signal.savgol_filter parity.

Parameters

Name

Type

Default

window_length

int

5

polyorder

int

2

deriv

int

0

delta

float

1.0

mode

str

'mirror'

cval

float

0.0

Explanations

Bibliographic source

Savitzky, A. & Golay, M. J. E. (1964). Smoothing and Differentiation of Data by Simplified Least Squares Procedures. Analytical Chemistry 36(8), 1627–1639.

Mathematical principle

Within a sliding window of odd length \(w\), a polynomial of order \(p\) is fit by least squares; the smoothed value (or its \(d\)-th derivative) at the window centre is a fixed linear combination of the windowed points. Because the convolution coefficients are precomputed, SG simultaneously denoises and differentiates while preserving peak shape far better than a moving average.

Implementation

C ABI n4m_pp_savgol_* in libn4m (create / apply / destroy lifecycle), wrapped by n4m.sklearn.SavitzkyGolay. The same numerical kernel backs every language binding.

Usage

from n4m.sklearn import SavitzkyGolay
op = SavitzkyGolay()
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).

BackendParity50×250 (ms)250×50 (ms)
C++ native · libn4m
pls4all.cpp.blas✓ ref
pls4all.cpp.blas+omp✓ ref0.02 ms0.07 ms
pls4all.cpp.omp✓ ref
pls4all.cpp.ref✓ ref0.02 ms🏆0.07 ms
Python · pls4all
pls4all.python✓ bind0.02 ms0.03 ms🏆
pls4all.sklearn✓ bind0.02 ms0.04 ms
Python · external
nirs4all⇄ +5e-020.29 ms0.34 ms
ref.python_numpysource

See also: methods index · interactive dashboard