pp_arpls — Ar P L S

Group: Baseline correction · Binding: n4m.sklearn.ArPLS · C ABI: n4m_pp_arpls_*

Description

Asymmetrically reweighted penalized least squares.

Parameters

Name

Type

Default

lam

float

100000.0

max_iter

int

50

tol

float

0.001

Explanations

Bibliographic source

Standard spectroscopic operator — see the nirs4all preprocessing / augmentation handbook and the cited literature within the binding docstring.

Mathematical principle

Asymmetrically reweighted penalized least squares.

Implementation

C ABI n4m_pp_arpls_* in libn4m (create / apply / destroy lifecycle), wrapped by n4m.sklearn.ArPLS. The same numerical kernel backs every language binding.

Usage

from n4m.sklearn import ArPLS
op = ArPLS()
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✓ ref3.96 ms3.35 ms
pls4all.cpp.omp✓ ref
pls4all.cpp.ref✓ ref3.34 ms3.67 ms
Python · pls4all
pls4all.python✓ bind2.38 ms2.64 ms
pls4all.sklearn✓ bind2.36 ms🏆2.62 ms🏆
Python · external
nirs4allsource20.0 ms87.2 ms
ref.python_numpysource

See also: methods index · interactive dashboard