pp_rnv — Robust Normal Variate (RNV)¶
Group: Preprocessing · Binding: n4m.sklearn.RNV · C ABI: n4m_pp_rnv_*
Description¶
Robust SNV using median + k * MAD.
Parameters¶
Name |
Type |
Default |
|---|---|---|
|
|
|
|
|
|
|
|
|
Explanations¶
Bibliographic source¶
Guo, Q., Wu, W. & Massart, D. L. (1999). The robust normal variate transform for pattern recognition with near-infrared data. Analytica Chimica Acta 382(1–2), 87–103.
Mathematical principle¶
A median/IQR analogue of SNV: each spectrum is corrected as \((\mathbf{x}_i - \mathrm{median}(\mathbf{x}_i)) / \mathrm{IQR}(\mathbf{x}_i)\). Replacing the mean and standard deviation with robust location/scale estimators makes the normalisation insensitive to a small number of strong absorption bands or outlying channels.
Implementation¶
C ABI n4m_pp_rnv_* in libn4m (create / apply / destroy lifecycle), wrapped by n4m.sklearn.RNV. The same numerical kernel backs every language binding.
Usage¶
from n4m.sklearn import RNV
op = RNV()
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).
| Backend | Parity | 50×250 (ms) | 250×50 (ms) |
|---|---|---|---|
| C++ native · libn4m | |||
pls4all.cpp.blas | ✓ ref | — | — |
pls4all.cpp.blas+omp | ✓ ref | 0.20 ms | 0.22 ms |
pls4all.cpp.omp | ✓ ref | — | — |
pls4all.cpp.ref | ✓ ref | 0.21 ms | 0.23 ms |
| Python · pls4all | |||
pls4all.python | ✓ bind | 0.19 ms | 0.20 ms |
pls4all.sklearn | ✓ bind | 0.19 ms🏆 | 0.19 ms🏆 |
| Python · external | |||
nirs4all | source | 0.34 ms | 0.37 ms |
ref.python_numpy | source | — | — |
See also: methods index · interactive dashboard