n2p2 - The neural network potential package, version midifed by A.R. Allouche

This repository provides ready-to-use software for high-dimensional neural network potentials in computational physics and chemistry. Added by A.R. Allouche :

• NN for partial charges and dipoles
• High derivatives of energies (2, 3, 4)
• dftd3 grimme correction to energie

Documentation

Build your own documentation

It is also possible to build your own documentation for offline reading. Install the above dependencies, change to the src directory and try to build the documentation:

source env.sh
cd dftd3-lib-0.9
make
cd ..
cd src
make doc

If the build process succeeds you can browse through the documentation starting from the main page in:

doc/html/index.html

Authors

• Andreas Singraber (University of Vienna)
• Abdulrahman Allouche (Lyon 1 University)

License

This software is licensed under the GNU General Public License version 3 or any later version (GPL-3.0-or-later).