DeePMD-kit v2: A software package for deep potential models

DeePMD-kit is a powerful open-source software package that facilitates molecular dynamics simulations using machine learning potentials (MLP) known as Deep Potential (DP) models. This package, which was released in 2017, has been widely used in the fields of physics, chemistry, biology, and material science for studying atomistic systems. The current version of DeePMD-kit offers numerous advanced features such as DeepPot-SE, attention-based and hybrid descriptors, the ability to fit tensile properties, type embedding, model deviation, Deep Potential - Range Correction (DPRc), Deep Potential Long Range (DPLR), GPU support for customized operators, model compression, non-von Neumann molecular dynamics (NVNMD), and improved usability, including documentation, compiled binary packages, graphical user interfaces (GUI), and application programming interfaces (API). This article presents an overview of the current major version of the DeePMD-kit package, highlighting its features and technical details. Additionally, the article benchmarks the accuracy and efficiency of different models and discusses ongoing developments.

[1]  Hui Li,et al.  A New Machine Learning Approach Based on Range Corrected Deep Potential Model for Efficient Vibrational Frequency Computation , 2023, 2303.15969.

[2]  Timothy J. Giese,et al.  Modern semiempirical electronic structure methods and machine learning potentials for drug discovery: Conformers, tautomers, and protonation states. , 2023, The Journal of chemical physics.

[3]  F. Paesani,et al.  A "short blanket" dilemma for a state-of-the-art neural network potential for water: Reproducing experimental properties or the physics of the underlying many-body interactions? , 2023, The Journal of chemical physics.

[4]  Timothy J. Giese,et al.  QDπ: A Quantum Deep Potential Interaction Model for Drug Discovery. , 2023, Journal of chemical theory and computation.

[5]  Yu Cao,et al.  Solvation structures of calcium and magnesium ions in water with the presence of hydroxide: a study by deep potential molecular dynamics. , 2023, Physical Chemistry, Chemical Physics - PCCP.

[6]  Kristof T. Schütt,et al.  SchNetPack 2.0: A neural network toolbox for atomistic machine learning. , 2022, The Journal of chemical physics.

[7]  Benjamin P. Pritchard,et al.  SPICE, A Dataset of Drug-like Molecules and Peptides for Training Machine Learning Potentials , 2022, Scientific Data.

[8]  Zhi X. Chen,et al.  Numerical investigation of azimuthal thermoacoustic instability in a gas turbine model combustor , 2021, Fuel.

[9]  E. Weinan,et al.  DeePKS-kit: A package for developing machine learning-based chemically accurate energy and density functional models , 2020, Comput. Phys. Commun..

[10]  M. de Koning,et al.  Plastic deformation of superionic water ices. , 2022, Proceedings of the National Academy of Sciences of the United States of America.

[11]  Jun Cheng,et al.  Resolving the odd-even oscillation of water dissociation at rutile TiO2(110)-water interface by machine learning accelerated molecular dynamics. , 2022, Journal of Chemical Physics.

[12]  Linfeng Zhang,et al.  DPA-1: Pretraining of Attention-based Deep Potential Model for Molecular Simulation , 2022, ArXiv.

[13]  D. York,et al.  Combined QM/MM, Machine Learning Path Integral Approach to Compute Free Energy Profiles and Kinetic Isotope Effects in RNA Cleavage Reactions. , 2022, Journal of chemical theory and computation.

[14]  D. Laage,et al.  Acids at the Edge: Why Nitric and Formic Acid Dissociations at Air-Water Interfaces Depend on Depth and on Interface Specific Area. , 2022, Journal of the American Chemical Society.

[15]  Siddarth K. Achar,et al.  Combined Deep Learning and Classical Potential Approach for Modeling Diffusion in UiO-66. , 2022, Journal of chemical theory and computation.

[16]  T. Ala‐Nissila,et al.  GPUMD: A package for constructing accurate machine-learned potentials and performing highly efficient atomistic simulations. , 2022, The Journal of chemical physics.

[17]  Jie Liu,et al.  Accurate and efficient molecular dynamics based on machine learning and non von Neumann architecture , 2022, npj Computational Materials.

[18]  Zachary W. Ulissi,et al.  GemNet-OC: Developing Graph Neural Networks for Large and Diverse Molecular Simulation Datasets , 2022, Trans. Mach. Learn. Res..

[19]  A. Yonezu,et al.  Development of neural network potential for MD simulation and its application to TiN , 2022, Computational Materials Science.

[20]  P. Debenedetti,et al.  Homogeneous ice nucleation in an ab initio machine-learning model of water , 2022, Proceedings of the National Academy of Sciences of the United States of America.

[21]  R. Car,et al.  Viscosity in water from first-principles and deep-neural-network simulations , 2022, npj Computational Materials.

[22]  K. Luo,et al.  Exploring Complex Reaction Networks Using Neural Network-Based Molecular Dynamics Simulation. , 2022, The journal of physical chemistry letters.

[23]  M. Klein,et al.  Dissolving salt is not equivalent to applying a pressure on water , 2022, Nature communications.

[24]  Weile Jia,et al.  Extending the limit of molecular dynamics with ab initio accuracy to 10 billion atoms , 2022, PPoPP.

[25]  K. Ho,et al.  A deep learning interatomic potential developed for atomistic simulation of carbon materials , 2022, Carbon.

[26]  W. E,et al.  A deep potential model with long-range electrostatic interactions. , 2021, The Journal of chemical physics.

[27]  Yaolong Zhang,et al.  REANN: A PyTorch-based end-to-end multi-functional deep neural network package for molecular, reactive, and periodic systems. , 2021, The Journal of chemical physics.

[28]  Linfeng Zhang,et al.  A tungsten deep neural-network potential for simulating mechanical property degradation under fusion service environment , 2021, Nuclear Fusion.

[29]  Steven J. Plimpton,et al.  LAMMPS - A flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales , 2021, Computer Physics Communications.

[30]  Linfeng Zhang,et al.  DP Compress: A Model Compression Scheme for Generating Efficient Deep Potential Models. , 2021, Journal of chemical theory and computation.

[31]  Toshiki Kataoka,et al.  Towards universal neural network potential for material discovery applicable to arbitrary combination of 45 elements , 2021, Nature Communications.

[32]  Jiaxin Zheng,et al.  Self‐Healing Mechanism of Lithium in Lithium Metal , 2021, Advanced science.

[33]  Yong Xu,et al.  Deep-learning density functional theory Hamiltonian for efficient ab initio electronic-structure calculation , 2021, Nature Computational Science.

[34]  F. Benistant,et al.  Efficient and accurate atomistic modeling of dopant migration using deep neural network , 2022, Materials Science in Semiconductor Processing.

[35]  Linfeng Zhang,et al.  Specialising neural network potentials for accurate properties and application to the mechanical response of titanium , 2021, npj Computational Materials.

[36]  Ming Hu,et al.  Thermodynamic and Transport Properties of LiF and FLiBe Molten Salts with Deep Learning Potentials. , 2021, ACS applied materials & interfaces.

[37]  T. Beck,et al.  Condensed Phase Water Molecular Multipole Moments from Deep Neural Network Models Trained on Ab Initio Simulation Data. , 2021, The journal of physical chemistry letters.

[38]  Shi Liu,et al.  Accurate force field of two-dimensional ferroelectrics from deep learning , 2021, Physical Review B.

[39]  R. Car,et al.  Heat transport in liquid water from first-principles and deep neural network simulations , 2021, Physical Review B.

[40]  Mario Barbatti,et al.  Choosing the right molecular machine learning potential , 2021, Chemical science.

[41]  Siddarth K. Achar,et al.  Efficiently Trained Deep Learning Potential for Graphane , 2021, Journal of Physical Chemistry C.

[42]  Z. Wen,et al.  2D Heterostructure of Amorphous CoFeB Coating Black Phosphorus Nanosheets with Optimal Oxygen Intermediate Absorption for Improved Electrocatalytic Water Oxidation. , 2021, ACS nano.

[43]  David T. Limmer,et al.  Learning intermolecular forces at liquid-vapor interfaces. , 2021, The Journal of chemical physics.

[44]  Y. Shao,et al.  Machine-Learning-Assisted Free Energy Simulation of Solution-Phase and Enzyme Reactions. , 2021, Journal of chemical theory and computation.

[45]  Pavlo O. Dral,et al.  MLatom 2: An Integrative Platform for Atomistic Machine Learning , 2021, Topics in Current Chemistry.

[46]  Lisandro Dalcin,et al.  mpi4py: Status Update After 12 Years of Development , 2021, Computing in Science & Engineering.

[47]  Jinyun Liu,et al.  Machine learning builds full-QM precision protein force fields in seconds. , 2021, Briefings in bioinformatics.

[48]  O. Isayev,et al.  Development of Multimodal Machine Learning Potentials: Toward a Physics-Aware Artificial Intelligence. , 2021, Accounts of chemical research.

[49]  D. York,et al.  Development of Range-Corrected Deep Learning Potentials for Fast, Accurate Quantum Mechanical/Molecular Mechanical Simulations of Chemical Reactions in Solution. , 2021, Journal of chemical theory and computation.

[50]  David T. Limmer,et al.  Reactive uptake of N2O5 by atmospheric aerosol is dominated by interfacial processes , 2021, Science.

[51]  W. E,et al.  Phase Diagram of a Deep Potential Water Model. , 2021, Physical review letters.

[52]  Michael Gastegger,et al.  Machine Learning Force Fields , 2020, Chemical reviews.

[53]  D. Matteson,et al.  PyXtal_FF: a python library for automated force field generation , 2020, Mach. Learn. Sci. Technol..

[54]  Alexander V. Shapeev,et al.  The MLIP package: moment tensor potentials with MPI and active learning , 2020, Mach. Learn. Sci. Technol..

[55]  E Weinan,et al.  86 PFLOPS Deep Potential Molecular Dynamics simulation of 100 million atoms with ab initio accuracy , 2020, Comput. Phys. Commun..

[56]  Zachary D. Hood,et al.  Fast Na Diffusion and Anharmonic Phonon Dynamics in Superionic Na3PS4 , 2021, Energy & Environmental Science.

[57]  B. Santra,et al.  Isotope effects in molecular structures and electronic properties of liquid water via deep potential molecular dynamics based on the SCAN functional , 2020, Physical Review B.

[58]  Yutong Lu,et al.  A DFT accurate machine learning description of molten ZnCl2 and its mixtures: 1. Potential development and properties prediction of molten ZnCl2 , 2020 .

[59]  Linfeng Zhang,et al.  Crystal Structure Prediction of Binary Alloys via Deep Potential , 2020, Frontiers in Chemistry.

[60]  M. Parrinello,et al.  Using metadynamics to build neural network potentials for reactive events: the case of urea decomposition in water , 2020, 2011.11455.

[61]  Jinzhe Zeng,et al.  Complex reaction processes in combustion unraveled by neural network-based molecular dynamics simulation , 2020, Nature Communications.

[62]  Wenshuo Liang,et al.  Molecular Dynamics Simulations of Molten Magnesium Chloride Using Machine‐Learning‐Based Deep Potential , 2020, Advanced Theory and Simulations.

[63]  A. Rempel,et al.  Deep machine learning interatomic potential for liquid silica. , 2020, Physical review. E.

[64]  R. Car,et al.  Hydrogen Dynamics in Supercritical Water Probed by Neutron Scattering and Computer Simulations. , 2020, The journal of physical chemistry letters.

[65]  Weihua Hu,et al.  The Open Catalyst 2020 (OC20) Dataset and Community Challenges , 2020, ACS Catalysis.

[66]  A. Henry,et al.  A deep neural network interatomic potential for studying thermal conductivity of β-Ga2O3 , 2020 .

[67]  T. Carrington,et al.  Neural Network Potential Energy Surfaces for Small Molecules and Reactions. , 2020, Chemical reviews.

[68]  P. Debenedetti,et al.  Signatures of a liquid–liquid transition in an ab initio deep neural network model for water , 2020, Proceedings of the National Academy of Sciences.

[69]  B. Santra,et al.  Isotope effects in x-ray absorption spectra of liquid water , 2020, Physical Review B.

[70]  Linfeng Zhang,et al.  Exploring the Chemical Space of Linear Alkane Pyrolysis via Deep Potential GENerator , 2020, Energy & Fuels.

[71]  Weinan E,et al.  Deep neural network for the dielectric response of insulators , 2020 .

[72]  Jaime Fern'andez del R'io,et al.  Array programming with NumPy , 2020, Nature.

[73]  Yanchun Zhou,et al.  Grain boundary strengthening in ZrB2 by segregation of W: Atomistic simulations with deep learning potential , 2020 .

[74]  P. Piaggi,et al.  Ab initio phase diagram and nucleation of gallium , 2020, Nature Communications.

[75]  Zachary L Glick,et al.  AP-Net: An atomic-pairwise neural network for smooth and transferable interaction potentials. , 2020, The Journal of chemical physics.

[76]  Justin S. Smith,et al.  TorchANI: A Free and Open Source PyTorch-Based Deep Learning Implementation of the ANI Neural Network Potentials , 2020, J. Chem. Inf. Model..

[77]  E Weinan,et al.  Pushing the Limit of Molecular Dynamics with Ab Initio Accuracy to 100 Million Atoms with Machine Learning , 2020, SC20: International Conference for High Performance Computing, Networking, Storage and Analysis.

[78]  Yanchun Zhou,et al.  Theoretical prediction on thermal and mechanical properties of high entropy (Zr0.2Hf0.2Ti0.2Nb0.2Ta0.2)C by deep learning potential , 2020 .

[79]  R. Car,et al.  Raman spectrum and polarizability of liquid water from deep neural networks. , 2020, Physical chemistry chemical physics : PCCP.

[80]  Zhilong Wang,et al.  Combining the Fragmentation Approach and Neural Network Potential Energy Surfaces of Fragments for Accurate Calculation of Protein Energy. , 2020, The journal of physical chemistry. B.

[81]  R. Car,et al.  Free energy of proton transfer at the water–TiO2 interface from ab initio deep potential molecular dynamics† , 2020, Chemical science.

[82]  Frank Noé,et al.  Machine learning for molecular simulation , 2019, Annual review of physical chemistry.

[83]  E Weinan,et al.  DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models , 2019, Comput. Phys. Commun..

[84]  T. Luo,et al.  A unified deep neural network potential capable of predicting thermal conductivity of silicon in different phases , 2019, 1912.05044.

[85]  Mohan Chen,et al.  Structure and dynamics of warm dense aluminum: a molecular dynamics study with density functional theory and deep potential , 2019, Journal of physics. Condensed matter : an Institute of Physics journal.

[86]  Seungwu Han,et al.  SIMPLE-NN: An efficient package for training and executing neural-network interatomic potentials , 2019, Comput. Phys. Commun..

[87]  Linfeng Zhang,et al.  Warm dense matter simulation via electron temperature dependent deep potential molecular dynamics , 2019, 1909.00231.

[88]  Bin Jiang,et al.  Embedded Atom Neural Network Potentials: Efficient and Accurate Machine Learning with a Physically Inspired Representation. , 2019, The journal of physical chemistry letters.

[89]  W. E,et al.  Isotope effects in liquid water via deep potential molecular dynamics , 2019, Molecular Physics.

[90]  Linfeng Zhang,et al.  Deep learning inter-atomic potential model for accurate irradiation damage simulations , 2019, Applied Physics Letters.

[91]  Markus Meuwly,et al.  PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments, and Partial Charges. , 2019, Journal of chemical theory and computation.

[92]  Christoph Dellago,et al.  Library-Based LAMMPS Implementation of High-Dimensional Neural Network Potentials. , 2019, Journal of chemical theory and computation.

[93]  Klaus-Robert Müller,et al.  sGDML: Constructing accurate and data efficient molecular force fields using machine learning , 2018, Comput. Phys. Commun..

[94]  E Weinan,et al.  Active Learning of Uniformly Accurate Inter-atomic Potentials for Materials Simulation , 2018, Physical Review Materials.

[95]  K-R Müller,et al.  SchNetPack: A Deep Learning Toolbox For Atomistic Systems. , 2018, Journal of chemical theory and computation.

[96]  Steven Vandenbrande,et al.  i-PI 2.0: A universal force engine for advanced molecular simulations , 2018, Comput. Phys. Commun..

[97]  Wei-Hai Fang,et al.  Deep Learning for Nonadiabatic Excited-State Dynamics. , 2018, The journal of physical chemistry letters.

[98]  Michele Parrinello,et al.  Silicon Liquid Structure and Crystal Nucleation from Ab Initio Deep Metadynamics. , 2018, Physical review letters.

[99]  Tai-Sung Lee,et al.  GPU-Accelerated Molecular Dynamics and Free Energy Methods in Amber18: Performance Enhancements and New Features , 2018, J. Chem. Inf. Model..

[100]  K. Butler,et al.  Machine learning for molecular and materials science , 2018, Nature.

[101]  Jun Li,et al.  Atomic Energies from a Convolutional Neural Network. , 2018, Journal of chemical theory and computation.

[102]  E Weinan,et al.  End-to-end Symmetry Preserving Inter-atomic Potential Energy Model for Finite and Extended Systems , 2018, NeurIPS.

[103]  Alexander Sergeev,et al.  Horovod: fast and easy distributed deep learning in TensorFlow , 2018, ArXiv.

[104]  K-R Müller,et al.  SchNet - A deep learning architecture for molecules and materials. , 2017, The Journal of chemical physics.

[105]  M Gastegger,et al.  wACSF-Weighted atom-centered symmetry functions as descriptors in machine learning potentials. , 2017, The Journal of chemical physics.

[106]  E Weinan,et al.  DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics , 2017, Comput. Phys. Commun..

[107]  E Weinan,et al.  Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics , 2017, Physical review letters.

[108]  Michael Walter,et al.  The atomic simulation environment-a Python library for working with atoms. , 2017, Journal of physics. Condensed matter : an Institute of Physics journal.

[109]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[110]  Vijay S. Pande,et al.  OpenMM 7: Rapid development of high performance algorithms for molecular dynamics , 2016, bioRxiv.

[111]  Klaus-Robert Müller,et al.  Machine learning of accurate energy-conserving molecular force fields , 2016, Science Advances.

[112]  Alexandre Tkatchenko,et al.  Quantum-chemical insights from deep tensor neural networks , 2016, Nature Communications.

[113]  J. Behler.  Perspective: Machine learning potentials for atomistic simulations. , 2016, The Journal of chemical physics.

[114]  Kevin Gimpel,et al.  Gaussian Error Linear Units (GELUs) , 2016 .

[115]  Timothy J. Giese,et al.  Ambient-Potential Composite Ewald Method for ab Initio Quantum Mechanical/Molecular Mechanical Molecular Dynamics Simulation. , 2016, Journal of chemical theory and computation.

[116]  Martín Ugarte,et al.  Foundations of JSON Schema , 2016, WWW.

[117]  Jian Sun,et al.  Identity Mappings in Deep Residual Networks , 2016, ECCV.

[118]  Berk Hess,et al.  GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers , 2015 .

[119]  Christopher D. Manning,et al.  Effective Approaches to Attention-based Neural Machine Translation , 2015, EMNLP.

[120]  Xiaohui Liu,et al.  Large-scale ab initio simulations based on systematically improvable atomic basis , 2015, 1503.00097.

[121]  Darrin M. York,et al.  Multipolar Ewald Methods, 1: Theory, Accuracy, and Performance , 2014, Journal of chemical theory and computation.

[122]  Timothy J. Giese,et al.  Multipolar Ewald Methods, 2: Applications Using a Quantum Mechanical Force Field , 2014, Journal of chemical theory and computation.

[123]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[124]  Andrew Collete,et al.  Python and HDF5: Unlocking Scientific Data , 2013 .

[125]  Yoshua Bengio,et al.  Deep Sparse Rectifier Neural Networks , 2011, AISTATS.

[126]  J. Behler.  Atom-centered symmetry functions for constructing high-dimensional neural network potentials. , 2011, The Journal of chemical physics.

[127]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[128]  R. Kondor,et al.  Gaussian approximation potentials: the accuracy of quantum mechanics, without the electrons. , 2009, Physical review letters.

[129]  Michele Parrinello,et al.  Generalized neural-network representation of high-dimensional potential-energy surfaces. , 2007, Physical review letters.

[130]  Darrin M York,et al.  An Efficient Linear-Scaling Ewald Method for Long-Range Electrostatic Interactions in Combined QM/MM Calculations. , 2005, Journal of chemical theory and computation.

[131]  George Bosilca,et al.  Open MPI: Goals, Concept, and Design of a Next Generation MPI Implementation , 2004, PVM/MPI.

[132]  William Gropp,et al.  MPICH2: A New Start for MPI Implementations , 2002, PVM/MPI.

[133]  Rajesh K. Gupta,et al.  YAML: a tool for hardware design visualization and capture , 2000, Proceedings 13th International Symposium on System Synthesis.

[134]  L. Dagum,et al.  OpenMP: an industry standard API for shared-memory programming , 1998 .

[135]  Steve Plimpton,et al.  Fast parallel algorithms for short-range molecular dynamics , 1993 .

[136]  O. Isayev,et al.  ANI-1: an extensible neural network potential with DFT accuracy at force fi eld computational cost † , 2017 .