Derived feature dataset, benchmark results, and reproducible methodology for subject-independent surface EMG gesture classification

This dataset contains derived results, benchmark tables, trained model artifacts, and a detailed reproducibility methodology, produced by a subject-disjoint machine-learning pipeline for surface electromyography (sEMG) gesture classification, applied to NinaPro Database 2 (40 subjects, 12 channels, 49 movements plus rest, 692,276 labelled 200 ms windows). NinaPro DB2 itself is a pre-existing, publicly available third-party dataset and is not redistributed here; every file in this deposit is an original output of the authors' own analysis. The dataset includes: held-out and 40-fold Leave-One-Subject-Out (LOSO) cross-validation results for three tuned gradient-boosted classifiers (CatBoost, XGBoost, LightGBM); per-subject LOSO ranking and calibration statistics; global feature-importance rankings from two independent methods (SHAP and permutation importance) plus a feature-level stability analysis across subjects; feature- and channel-level ablation results quantifying the cost of removing top- vs. bottom-ranked features; embedded-deployment benchmarking results (stage-by-stage inference latency, model export size across four formats, and a direct test of post-training quantization applicability); the tuned CatBoost model exported in both native binary (.cbm) and ONNX formats; and a step-by-step methodology document (METHODOLOGY.md) sufficient for independent reproduction of every derived file from the source recordings. This is an updated, standalone deposit that supersedes an earlier, more limited version of this dataset (14 files, without the methodology document); see Related Works for the link between the two records. These derived data support independent verification and reuse of the reported classification performance, explainability cross-checks, and deployment feasibility claims, and allow reproduction of every file from the source recordings via the accompanying methodology document, without requiring re-derivation of the analysis pipeline from scratch.