Processed-feature LOSO benchmark data and code for sEMG protocol-state discriminability in a cable-driven walking-assistance suit

This record provides the processed-feature datasets, source data, LOSO split manifest, scoring scripts, and analysis code supporting the revised manuscript, "A reproducible processed-feature LOSO benchmark for sEMG protocol-state discriminability in a cable-driven walking-assistance suit." The package is intended to support reproducibility at the processed-feature/source-data level. Raw 10-channel sEMG time-series data are not included because participant-consent and IRB-approved ethics conditions do not permit public redistribution of raw physiological recordings. Raw data may be available from the corresponding author on reasonable request for non-commercial academic research, subject to institutional approval, IRB/consent constraints, and a Data Use Agreement.