Paper:

• Jakub Martinka, Mikołaj Martyka, Jiří Pittner* and Pavlo O. Dral*. Flexible Framework for Surface Hopping: From Hybrid Schemes for Machine Learning to Benchmarkable Nonadiabatic Dynamics. Journal of Chemical Theory and Computation 2026 22 (5), 2467-2479. DOI: 10.1021/acs.jctc.5c02137

Preprint on: 10.48550/arXiv.2512.19152

• This repository contains a Jupyter notebook with Supporting Information (SI.ipynb), examples/ containing input files and scripts to run surface hopping simulations and res/ with resulting figures.

Run ansep.py to create TRAJs or create pop.txt, run surface hopping by run.py, data/ contains initial conditions and QC inputs and/or ML models.

Structure:
examples/
│
├── CNH4p_CASSCF_FSSH
├── CNH4p_CASSCF_LZSH
├── CNH4p_CASSCF_TDBA
├── CNH4p_MRCI_FSSH
│
├── FERRO_WIRE_FSSH
├── FERRO_WIRE_LZSH
├── FERRO_WIRE_TDBA
├── FERRO_WIRE_TDBA_0.5eV
│
├── FULVENE_FSSH
├── FULVENE_FSSH_MLQM
├── FULVENE_FSSH_MLQM_0.5eV
├── FULVENE_LZSH
├── FULVENE_TDBA
├── FULVENE_TDBA_0.5eV
└── FULVENE_TDBA_dgrad - rescale velocity $g_{ij}$
res/

Trajectories in H5MD format (exceeding 55 GB in total) can be provided upon request due to their size.