DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models

DP-GEN (Deep Generator) is a software written in Python, delicately designed to generate a deep learning based model of interatomic potential energy and force field. DP-GEN is dependent on DeePMD-kit. With highly scalable interface with common softwares for molecular simulation, DP-GEN is capable to automatically prepare scripts and maintain job queues on HPC machines (High Performance Cluster) and analyze results.

If you use this software in any publication, please cite:

Yuzhi Zhang, Haidi Wang, Weijie Chen, Jinzhe Zeng, Linfeng Zhang, Han Wang, and Weinan E, DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models, Computer Physics Communications, 2020, 253, 107206.

Highlighted features

• Accurate and efficient: DP-GEN is capable to sample more than tens of million structures and select only a few for first principles calculation. DP-GEN will finally obtain a uniformly accurate model.
• User-friendly and automatic: Users may install and run DP-GEN easily. Once successfully running, DP-GEN can dispatch and handle all jobs on HPCs, and thus there's no need for any personal effort.
• Highly scalable: With modularized code structures, users and developers can easily extend DP-GEN for their most relevant needs. DP-GEN currently supports for HPC systems (Slurm, PBS, LSF and cloud machines), Deep Potential interface with DeePMD-kit, MD interface with LAMMPS, Gromacs, AMBER, Calypso and ab-initio calculation interface with VASP, PWSCF, CP2K, SIESTA, Gaussian, Abacus, PWmat, etc. We're sincerely welcome and embraced to users' contributions, with more possibilities and cases to use DP-GEN.

Download and Install

DP-GEN only supports Python 3.8 and above.

One can download the source code of dpgen by

git clone https://github.com/deepmodeling/dpgen.git

then you may install DP-GEN easily by:

cd dpgen
pip install --user .

With this command, the dpgen executable is install to $HOME/.local/bin/dpgen. You may want to export the PATH by

export PATH=$HOME/.local/bin:$PATH

To test if the installation is successful, you may execute

dpgen -h

Workflows and usage

DP-GEN contains the following workflows:

• dpgen run: Main process of Deep Generator.
• Init: Generating initial data.
  • dpgen init_bulk: Generating initial data for bulk systems.
  • dpgen init_surf: Generating initial data for surface systems.
  • dpgen init_reaction: Generating initial data for reactive systems.
• dpgen simplify: Reducing the amount of existing dataset.
• dpgen autotest: Autotest for Deep Potential.

For detailed usage and parameters, read DP-GEN documentation.

Tutorials and examples

• Tutorials: basic tutorials for DP-GEN.
• Examples: input files in JSON format.
• Publications: Published research articles using DP-GEN.
• User guide: frequently asked questions listed in troubleshooting.

License

The project dpgen is licensed under GNU LGPLv3.0.

Contributing

DP-GEN is maintained by DeepModeling's developers. Contributors are always welcome.