filter_quality — Rule-based row-level spectral quality filter¶
Group: Sample / feature filters · C ABI: n4m_outlier_detection_spectral_quality_*
Description¶
Stateless row-level spectrum quality filter.
Parameters¶
Name |
Type |
Default |
|---|---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
API and bindings¶
C ABI (ABI 2): n4m_outlier_detection_spectral_quality_apply · n4m_outlier_detection_spectral_quality_create · n4m_outlier_detection_spectral_quality_destroy. Use the linked public header for the exact signature, configuration, and result handles.
Python (verified public re-export):
from n4m.outlier_detection import SpectralQualityFilter
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 conjunction of data-quality checks. It is an explicit internal quality-control rule set whose thresholds are user policy.
Mathematical principle¶
A row is kept only if every enabled rule passes: NaN fraction \(\le\) max_nan_ratio; no infinity when check_inf; exact-zero fraction \(\le\) max_zero_ratio; NaN-aware population variance \(\ge\) min_variance; and optional finite extrema within min_value/max_value. Fit is an idempotent no-op.
Appropriate uses¶
Rejecting empty, flat, non-finite, saturated, or out-of-range acquisitions before modeling.
Limits and validation¶
Thresholds are scale- and representation-specific. NaNs count as nonzero for the zero-ratio rule, and a row can pass these syntactic checks while remaining chemically implausible.
Implementation¶
Python role API n4m.outlier_detection.SpectralQualityFilter; ABI 2 family n4m_outlier_detection_spectral_quality_{create,apply,destroy}.
The ABI-2 implementation is the n4m_outlier_detection_spectral_quality_* lifecycle in libn4m.
Sources and provenance¶
https://github.com/GBeurier/nirs4all-methods/blob/main/cpp/src/core/filters/spectral_quality.h
See also: methods index · interactive dashboard