aug_spline_x_simplify — Cubic-spline reconstruction from sparse x controls

Group: Augmentation · C ABI: n4m_augmentation_spline_x_simplification_*

Description

Simplify each spectrum via a cubic B-spline through a random control subset on the x-axis (nirs4all Spline_X_Simplification). spline_points <= 0 selects the reference default of n_features // 4. Mirrors the reference defaults (uniform=False).

Parameters

Name

Type

Default

spline_points

int

-1

uniform

bool

False

rng

Optional[PCG64]

None

seed

int

0

API and bindings

C ABI (ABI 2): n4m_augmentation_spline_x_simplification_apply · n4m_augmentation_spline_x_simplification_create · n4m_augmentation_spline_x_simplification_destroy. Use the linked public header for the exact signature, configuration, and result handles.

Python (verified public re-export):

from n4m.augmentation.splines import SplineXSimplificationAugmenter

Source signature: SplineXSimplificationAugmenter(spline_points: int = -1, uniform: bool = False, rng: Optional[PCG64] = None, seed: int = 0).

R: no current source-verified entry point was found for this catalog method.

MATLAB / Octave: no current source-verified entry point was found for this catalog method.

Explanations

Bibliographic source

No canonical paper defines this augmentation heuristic. It uses a classical interpolating B-spline but the control subset and parity rules are implementation-specific.

Mathematical principle

Keep spline_points channel/value controls (default \(p/4\)), selected uniformly or by PCG64 sampling without replacement. Fit an interpolating not-a-knot cubic B-spline through those controls and evaluate it at all original channels.

Appropriate uses

Testing robustness to reduced effective spectral resolution and sparse sampling.

Limits and validation

Fine features between controls are irrecoverable and cubic interpolation may overshoot. This x variant does not apply the curve variant’s final unique step in uniform mode.

Implementation

Python role API n4m.augmentation.splines.SplineXSimplificationAugmenter; ABI 2 family n4m_augmentation_spline_x_simplification_{create,apply,destroy}.

The ABI-2 implementation is the n4m_augmentation_spline_x_simplification_* lifecycle in libn4m.

Sources and provenance

https://github.com/GBeurier/nirs4all-methods/blob/main/cpp/src/core/augmentation/splines/spline_simplify_common.h

See also: methods index · interactive dashboard