Data and Code — Design Rules for Carbon Nanotube Yarns: Wall-Count-Dependent Interwall Crosslinking Revealed by Machine-Learning-Assisted Molecular Dynamics

Data and code supporting "Design Rules for Carbon Nanotube Yarns: Wall-Count-Dependent Interwall Crosslinking Revealed by Machine-Learning-Assisted Molecular Dynamics." Contains the molecular dynamics structure-generation scripts, LAMMPS run decks, post-processing scripts, the Bayesian-optimization (kriging + GA) input decks, an example surrogate-training sample, and the optimization-trajectory results for all three campaigns. A README documents the full pipeline and reproduction order.