`pp_baseline` — Training-set column mean centering¶

Group: Preprocessing · C ABI: n4m_transform_baseline_center_*

Description¶

Column-mean baseline centering.

Parameters¶

No constructor parameters.

API and bindings¶

C ABI (ABI 2): n4m_transform_baseline_center_create · n4m_transform_baseline_center_destroy · n4m_transform_baseline_center_fit · n4m_transform_baseline_center_inverse_transform · n4m_transform_baseline_center_is_fitted · n4m_transform_baseline_center_transform. Use the linked public header for the exact signature, configuration, and result handles.

Python (verified public re-export):

from n4m.transform.scaling import BaselineCenter

Source signature: BaselineCenter().

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 single canonical paper: this is the standard centering operation used by PCA, PLS, and linear models. The implementation source is the normative definition.

Mathematical principle¶

Fit stores \(\mu_j=n^{-1}\sum_i X_{ij}\) for every wavelength; transform returns \(X_{ij}-\mu_j\) and inverse transform adds the same training mean. It is feature centering across samples, not row-wise spectral baseline subtraction.

Appropriate uses¶

Preparing calibration matrices for covariance-based models and reproducing the training coordinate system on validation or prediction data.

Limits and validation¶

It only removes a constant offset per feature and is stateful: fitting before a data split leaks validation information. It does not correct per-spectrum drift.

Implementation¶

n4m.transform.scaling.BaselineCenter uses n4m_transform_baseline_center_*; scaling/baseline.c stores and reuses the training column means.

The ABI-2 implementation is the n4m_transform_baseline_center_* lifecycle in libn4m.

Sources and provenance¶

https://github.com/GBeurier/nirs4all-methods/blob/main/cpp/src/core/preprocessing/scaling/baseline.c

See also: methods index · interactive dashboard

pp_baseline — Training-set column mean centering¶