DeePMD-kit v2: A software package for deep potential models
DeePMD-kit v2: программный пакет для моделей глубокого потенциала
2023-08-01
SCID: 54.1/cytjszw4
Discuss with AI
DP-range correctionDeep Potential (DP) modelsDeepPot-SEattention-based descriptorsmodel compression
Figures from the paper
Abstract (AI)
DeePMD-kit is a powerful open-source software package that facilitates molecular dynamics simulations using machine learning potentials 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, DP-range correction, DP long range, graphics processing unit support for customized operators, model compression, non-von Neumann molecular dynamics, and improved usability, including documentation, compiled binary packages, graphical user interfaces, and application programming interfaces. This article presents an overview of the current major version of the DeePMD-kit package, highlighting its features and technical details. Additionally, this article presents a comprehensive procedure for conducting molecular dynamics as a representative application, benchmarks the accuracy and efficiency of different models, and discusses ongoing developments.
Key Findings
1
DeePMD-kit is an open-source software package enabling molecular dynamics with Deep Potential (DP) machine learning potentials since 2017.
2
New capabilities include fitting tensile properties, model deviation estimation, DP-range correction, and DP long range treatments.
3
The article provides a comprehensive MD workflow, benchmarks accuracy and efficiency of different DP models, and discusses ongoing developments.
4
The current major version adds many advanced features including DeepPot-SE, attention-based and hybrid descriptors, and type embedding.
5
The package supports GPU acceleration for customized operators, model compression, and non-von Neumann molecular dynamics.
6
Usability improvements include enhanced documentation, compiled binary packages, graphical user interfaces, and application programming interfaces.
Research Object
DeePMD-kit software package for Deep Potential (DP) models
Research Subject
Features, technical implementation, and performance (accuracy and efficiency) of DeePMD-kit for enabling molecular dynamics simulations with Deep Potential models, including advanced descriptors, long-range corrections, GPU/custom operator support, model compression, usability, and benchmarking
Publication Details
Publication Date
2023-08-01
Journal
Publisher
ISSN
Cited by
596
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai7
Adam: A Method for Stochastic Optimization2014
Exploiting Generative AI to Scale up Intelligent Tutoring Systems2023
Array programming with NumPy2020
GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers2015
LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales2021
Effective Approaches to Attention-based Neural Machine Translation2015
The atomic simulation environment—a Python library for working with atoms2017