Skip to content

matlab/matlab-agentic-toolkit / skills-catalog/ai-and-statistics/matlab-engineer-tabular-features

matlab-engineer-tabular-features

Unverified

Use when engineering or selecting the best features for single-response classification or regression in MATLAB, whatever the data's modality — for non-tabular data it routes extraction to a domain skill, then selects, assesses, and delivers on the resulting table. Not for multi-response problems, model training, or raw data acquisition.

FileKindSize
SKILL.mdskill md18.9 KB
manifest.yamlconfig570 B
references/assess.mdmarkdown16.7 KB
references/deliver.mdmarkdown13.0 KB
references/domain-routing.mdmarkdown18.1 KB
references/feature-pool.mdmarkdown10.9 KB
references/intake.mdmarkdown16.9 KB
references/internal-helpers.mdmarkdown12.2 KB
references/provider-protocol.mdmarkdown12.3 KB
references/report.mdmarkdown9.7 KB
references/reproduce-pool.mdmarkdown7.1 KB
references/select.mdmarkdown7.1 KB
scripts/BordaVoter.mother2.3 KB
scripts/ConsensusVoter.mother2.5 KB
scripts/MajorityVoter.mother2.0 KB
scripts/RRFVoter.mother1.9 KB
scripts/assessGenerationStability.mother11.9 KB
scripts/assessKFold.mother20.6 KB
scripts/assessSelectionStability.mother9.8 KB
scripts/baselineComparison.mother18.9 KB
scripts/buildRecipe.mother8.2 KB
scripts/chooseNumBins.mother5.0 KB
scripts/chooseStandardization.mother4.5 KB
scripts/computeNSelect.mother2.9 KB
scripts/createConsensusVoter.mother2.2 KB
scripts/describeFeatures.mother5.1 KB
scripts/expandBlockMap.pbinary440 B
scripts/featureSetQuality.mother12.3 KB
scripts/generateFeatureReport.mother67.2 KB
scripts/generateFeatureReportPdf.mother64.9 KB
scripts/generateFeatures.mother14.6 KB
scripts/hasReportGenerator.mother630 B
scripts/isDomainRecipe.mother1.4 KB
scripts/isStringEmpty.mother577 B
scripts/lambdaSequence.pbinary637 B
scripts/lassoFeatureRanking.pbinary971 B
scripts/macroF1Score.mother931 B
scripts/mrmrFeatureRanking.pbinary239 B
scripts/mutualInformationPanel.pbinary1.1 KB
scripts/ncaFeatureRanking.pbinary464 B
scripts/nogueiraStability.mother3.8 KB
scripts/oobFeatureRanking.pbinary414 B
scripts/plotSelectionConsensus.mother2.7 KB
scripts/plotSelectionDecision.mother6.3 KB
scripts/plotSelectionStability.mother10.2 KB
scripts/plotValidationBaseline.mother6.1 KB
scripts/printSummary.mother18.3 KB
scripts/profileForSplit.mother6.2 KB
scripts/rankAgreement.mother1.3 KB
scripts/rankFeatures.mother6.2 KB
scripts/reserveHoldoutForUser.mother5.5 KB
scripts/runConsensusSelection.mother9.4 KB
scripts/sampleSkewness.mother600 B
scripts/saveResults.mother2.1 KB
scripts/scoreFeaturePanel.pbinary1.6 KB
scripts/screenPredictors.mother13.3 KB
scripts/selectFeatures.mother2.3 KB
scripts/splitStrategy.mother5.4 KB
scripts/stratifiedDownsampleRows.mother2.2 KB
scripts/stratifiedSubsampleIndices.mother3.4 KB
scripts/transformFeatures.mother6.1 KB
scripts/validateRecipe.mother12.3 KB
scripts/writeInferenceScript.mother10.5 KB

scripts/computeNSelect.m

View on GitHub
function NSelect = computeNSelect(Scores, NCandidates)
%computeNSelect Compute number of features to select via score elbow detection.
%
%   NSelect = computeNSelect(Scores, NCandidates) finds the elbow in the sorted
%   consensus scores as the point of maximum BELOW-CHORD distance from the line
%   connecting the first and last points, clamped to the pool size NCandidates.
%   Only a convex curve -- one whose scores dip below that chord -- has a knee to
%   cut at; a linear decline (scores on the chord) or a concave one (scores above
%   it) has no below-chord point, so there is no elbow and the function keeps all
%   features. A signed distance is used deliberately: taking abs() would
%   manufacture a spurious "elbow" near the tail of a concave curve where none
%   exists.
%
%   Distance is measured as the VERTICAL drop from the chord down to each score,
%   not the perpendicular distance. The two are proportional by the constant
%   factor DX/LineMag (a property of the chord, identical for every point), so
%   they are maximized at the exact same feature -- the elbow is unchanged either
%   way. Vertical drop is chosen because it is what plotSelectionDecision draws to
%   illustrate the cut, so the metric and its picture are one and the same
%   quantity rather than merely proportional.
%
%   Inputs:
%     Scores      - (p,1) or (1,p) numeric, consensus scores (unsorted or sorted)
%     NCandidates - (1,1) positive integer, pool size (upper clamp)
%
%   Output:
%     NSelect - number of features to select (at least 2, at most NCandidates)

% Copyright 2026 The MathWorks, Inc.

    arguments
        Scores (:,1) double {mustBeNonempty}
        NCandidates (1,1) {mustBePositive, mustBeInteger}
    end

    Scores = sort(Scores, 'descend');
    p = numel(Scores);

    if p <= 3
        NSelect = min(p, NCandidates);
        return;
    end

    X1 = 1;       Y1 = Scores(1);
    X2 = p;       Y2 = Scores(end);
    DX = X2 - X1;
    DY = Y2 - Y1;

    if DX < eps
        NSelect = min(p, NCandidates);
        return;
    end

    % Signed vertical drop from the chord down to each score. With scores
    % sorted descending and the chord running from point 1 to point p, a point
    % that dips BELOW the chord (the convex knee we want) yields a positive
    % value; a point on the chord yields zero (linear, no knee); a point above
    % it yields a negative value (concave, no knee). Taking the max therefore
    % fires only when a genuine below-chord elbow exists. This is the same
    % quantity plotSelectionDecision draws as the vertical construction line.
    ChordY = Y1 + ((1:p)' - X1) / DX * DY;
    Distances = ChordY - Scores;

    [MaxDist, ElbowIdx] = max(Distances);

    if MaxDist < 1e-10
        % No point falls below the chord: linear or concave decline, no elbow.
        NSelect = min(p, NCandidates);
        return;
    end

    % A genuine convex elbow: it alone drives the count.
    NSelect = min(max(2, ElbowIdx), NCandidates);
end
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 466kUpdated Claude CodeCodexVerifiedInstall
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 466kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 319kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 319kUpdated Claude CodeCodexVerifiedInstall