rep_select — REP — Recursive Elimination of Predictors

Group: Variable selector · Registry tolerance: 1e-06

Description

REP-PLS repeated VIP selection (§18 Phase 5s)

From the pls4all.sklearn.REPSelector docstring:

REP-PLS — repeated VIP-thresholded variable selection.

Registry note — R plsVarSel::rep_pls repeated VIP-filtered selection. Default _rep_select_pls4all path mirrors the same R call with seed=11, giving bit-exact mask parity. The C++ step-count backward-elimination kernel is opt-in via legacy=True.

Parameters

Name

Type

Default

Notes

n_components

int

2

Number of latent components extracted (k).

n_steps

int

10

Number of elimination passes performed.

min_features

`int

None`

None

remove_count

int

1

Number of variables removed per REP step.

n_folds

int

3

Number of cross-validation folds used inside the selector.

rep_ratio

float

0.5

registry benchmark cell value

rep_vip_threshold

float

0.5

registry benchmark cell value

rep_repeats

int

3

registry benchmark cell value

Explanations

Bibliographic source

Mehmood, T., Liland, K. H., Snipen, L. & Sæbø, S. (2012). A review of variable selection methods in partial least squares regression. Chemometrics and Intelligent Laboratory Systems 118, 62–69. https://doi.org/10.1016/j.chemolab.2012.07.010 — same review as shaving_select; §3.3 Recursive elimination introduces the fixed-count variant implemented here.

Mathematical principle

REP removes a fixed count of features per recursive step (rather than a fraction as in shaving). At each step, sort features by absolute coefficient score, remove the \(m\) lowest, refit, record CV-RMSE. Return the subset with lowest CV-RMSE across all retained trajectories.

Useful when the analyst wants control over total iteration count: with \(m\) features removed per step, the process terminates in \(\lceil p / m \rceil\) iterations. Same intent as shaving but with linear instead of geometric decay.

Implementation

n4m_rep_select.

Usage

Every pls4all binding tab dispatches into the same C kernel; the external libraries listed at the bottom of the page are the parity references registered in benchmarks.parity_timing.registry. Switch tabs to read the same fit in your language. The R package now ships drop-in-compatible facades for the CRAN pls package (plsr, pcr, mvr) and for the mdatools::pls(x, y, ...) matrix idiom — those tabs appear only on the methods that have a meaningful equivalence.

pls4all bindings

/* C ABI — libn4m */
n4m_context_t* ctx = n4m_context_create();
n4m_config_t*  cfg = n4m_config_create();
n4m_method_result_t* res = NULL;
n4m_rep_select_fit(ctx, cfg, &x_view, &y_view, /* hyperparams */, &res);
/* … read coefficients / mask / scores via */
/* n4m_method_result_get_double_matrix / vector / scalar … */
n4m_method_result_destroy(res);
n4m_config_destroy(cfg);
n4m_context_destroy(ctx);

import pls4all
from pls4all._methods import rep_select_fit
with pls4all.Context() as ctx, pls4all.Config() as cfg:
    res = rep_select_fit(ctx, cfg, X, y, n_components=3)
# then: res.matrix("predictions"), res.matrix("coefficients"),
# res.vector("mask"), res.scalar("intercept"), …

from pls4all.sklearn import REPSelector
mdl = REPSelector(n_components=2, n_steps=10, min_features=None, remove_count=1, n_folds=3)
mdl.fit(X, y)
y_hat = mdl.predict(X_test)

library(pls4all)
# Unified low-level dispatcher (May 2026 R cleanup):
res <- pls4all_method("rep_select", X, y,
                      n_components = 3L, params = list(n_steps = 9L, min_features = 6L, remove_count = 5L, rep_ratio = 0.5, rep_vip_threshold = 0.5, rep_repeats = 3L))
# res is a named list with MethodResult arrays/scalars.
# selected_indices / top_k_intervals are 1-based.

res = pls4all.rep_select(X, y, 3);
% see header of bindings/matlab/+pls4all/rep_select.m for full
% parameter surface:
%   res = rep_select(X, Y, n_components, n_steps, min_features, remove_count)
yhat = predict(res, Xtest);

No idiomatic classdef wrapper — invoke pls4all.fit("rep_select", X, y, …) directly from the unified MEX factory.

Registry parity references 📐

  • 📐 ref.r_plsvarsel (R · r) — plsVarSel 0.10.0 · strict (rmse_rel ≤ 1e-06) — R plsVarSel::rep_pls — repeated VIP-thresholded variable selection.

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-06).

Rows tagged with 📐 are the canonical parity references for this method (declared in parity_timing.registry). C++ and external rows show reference parity; pls4all language bindings show binding parity against the C++ backend. Hover the icon for role and tolerance band.

BackendParity200×40 (ms)
C++ native · libn4m
pls4all.cpp.blas+omp✓ J 1.00605.7 ms
Python · pls4all
pls4all.python✓ J 1.00588.4 ms
pls4all.sklearn⇄ J 0.213.35 ms🏆
R · pls4all
pls4all.R⇄ J 0.215.16 ms
pls4all.R.formula⇄ J 0.216.34 ms
pls4all.R.mdatools⇄ J 0.216.53 ms
pls4all.R.pls⇄ J 0.216.07 ms
R · external
📐ref.r_plsvarselsource241.7 ms
BackendParity200×40 (ms)
C++ native · libn4m
pls4all.cpp.blas+omp✓ J 1.00589.6 ms
Python · pls4all
pls4all.python✓ J 1.00594.0 ms
pls4all.sklearn⇄ J 0.212.18 ms🏆
R · pls4all
pls4all.R⇄ J 0.215.01 ms
pls4all.R.formula⇄ J 0.215.61 ms
pls4all.R.mdatools⇄ J 0.215.95 ms
pls4all.R.pls⇄ J 0.215.87 ms
R · external
📐ref.r_plsvarselsource239.4 ms
BackendParity200×40 (ms)
C++ native · libn4m
pls4all.cpp.blas+omp✓ J 1.00611.9 ms
Python · pls4all
pls4all.python✓ J 1.00630.6 ms
pls4all.sklearn⇄ J 0.212.24 ms🏆
R · pls4all
pls4all.R⇄ J 0.214.92 ms
pls4all.R.formula⇄ J 0.216.09 ms
pls4all.R.mdatools⇄ J 0.215.95 ms
pls4all.R.pls⇄ J 0.215.79 ms
R · external
📐ref.r_plsvarselsource246.5 ms

See also: benchmark overview · methods index · interactive dashboard