In order to develop Neural Network potential-energy surfaces for a variety of system, we have developed our in-house NN code for high-dimensional systems called RuNNer, which was the first implementation of high-dimensional NN potentials. Much of the methodology has been originally developed with this code. The code is freely available under the GPL3 license. We currently host the source code in a repository at gitlab. If you are interested in obtaining a free copy of RuNNer, please write an email to joerg.behler@rub.de stating your name, institution, official institutional email address and your gitlab username so that we can provide you access.


RuNNer has the following features:
 

• unlimited number of degrees of freedom (atoms)
• training data can be obtained from arbitrary electronic structure methods and codes
• training using energies and forces
• periodic and non-periodic system
• several types of symmetry functions are available
• several types of activation functions are available
• arbitrary topology of the atomic neural networks
• provides energies and analytic derivatives (forces and stress tensor)

The methodology of RuNNer is published in the following papers:

J. Behler, and M. Parrinello, Phys. Rev. Lett. 98, 146401 (2007).
J. Behler, J. Chem. Phys. 134, 074106 (2011).
J. Behler, J. Phys.: Condens. Matter 26, 183001 (2014).
J. Behler, Int. J. Quantum Chem. 115, 1032 (2015)
J. Behler, Angew. Chem. Int. Ed. 56 (2017) 12828.

If you are interested in learning more about RuNNer, please feel free to contact us.

Coming soon:

We are currently preparing a new completely revised new release RuNNer2.0 with contributions by Alexander Knoll, Moritz Schäfer, Nikolas Lausch, Gunnar Schmitz, Moritz Gubler, Alea Tokita, Henry Wang, Richard Springborn, Marco Eckhoff, Redouan El Haouari and Jörg Behler. More information will follow soon.