pp_wavelet_denoise — Wavelet Denoise

Group: Augmentation · Binding: n4m.sklearn.WaveletDenoise · C ABI: n4m_pp_wavelet_denoise_*

Description

Multi-level DWT VisuShrink denoising.

Full binding docstring 
Multi-level DWT VisuShrink denoising.

Stateless: matches PyWavelets' ``waverec(threshold(coeffs))`` pipeline.

Parameters

Name

Type

Default

family

str

'db4'

mode

str

'periodization'

level

int

3

threshold_mode

str

'soft'

noise_estimator

str

'median'

Explanations

Bibliographic source

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

Mathematical principle

Multi-level DWT VisuShrink denoising.

Implementation

C ABI n4m_pp_wavelet_denoise_* in libn4m (create / apply / destroy lifecycle), wrapped by n4m.sklearn.WaveletDenoise. The same numerical kernel backs every language binding.

Usage

from n4m.sklearn import WaveletDenoise
op = WaveletDenoise()
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✓ ref0.13 ms0.12 ms
pls4all.cpp.omp✓ ref——
pls4all.cpp.ref✓ ref0.13 ms0.14 ms
Python · pls4all
pls4all.python✓ bind0.09 ms0.10 ms
pls4all.sklearn✓ bind0.09 ms🏆0.10 ms🏆
Python · external
nirs4all⇄ +1e-010.26 ms0.32 ms
ref.python_numpysource——

See also: methods index · interactive dashboard