DeePMD-kit’s documentation

DeePMD-kit is a package written in Python/C++, designed to minimize the effort required to build deep learning-based models of interatomic potential energy and force field and to perform molecular dynamics (MD). This brings new hopes to addressing the accuracy-versus-efficiency dilemma in molecular simulations. Applications of DeePMD-kit span from finite molecules to extended systems and from metallic systems to chemically bonded systems.

Important

The project DeePMD-kit is licensed under GNU LGPLv3.0. If you use this code in any future publications, please cite the following publications for general purpose:

• Han Wang, Linfeng Zhang, Jiequn Han, and Weinan E. “DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics.” Computer Physics Communications 228 (2018): 178-184.

• Jinzhe Zeng, Duo Zhang, Denghui Lu, Pinghui Mo, Zeyu Li, Yixiao Chen, Marián Rynik, Li’ang Huang, Ziyao Li, Shaochen Shi, Yingze Wang, Haotian Ye, Ping Tuo, Jiabin Yang, Ye Ding, Yifan Li, Davide Tisi, Qiyu Zeng, Han Bao, Yu Xia, Jiameng Huang, Koki Muraoka, Yibo Wang, Junhan Chang, Fengbo Yuan, Sigbjørn Løland Bore, Chun Cai, Yinnian Lin, Bo Wang, Jiayan Xu, Jia-Xin Zhu, Chenxing Luo, Yuzhi Zhang, Rhys E. A. Goodall, Wenshuo Liang, Anurag Kumar Singh, Sikai Yao, Jingchao Zhang, Renata Wentzcovitch, Jiequn Han, Jie Liu, Weile Jia, Darrin M. York, Weinan E, Roberto Car, Linfeng Zhang, Han Wang. “DeePMD-kit v2: A software package for Deep Potential models.” J. Chem. Phys., 159, 054801 (2023).

• Jinzhe Zeng, Duo Zhang, Anyang Peng, Xiangyu Zhang, Sensen He, Yan Wang, Xinzijian Liu, Hangrui Bi, Yifan Li, Chun Cai, Chengqian Zhang, Yiming Du, Jia-Xin Zhu, Pinghui Mo, Zhengtao Huang, Qiyu Zeng, Shaochen Shi, Xuejian Qin, Zhaoxi Yu, Chenxing Luo, Ye Ding, Yun-Pei Liu, Ruosong Shi, Zhenyu Wang, Sigbjørn Løland Bore, Junhan Chang, Zhe Deng, Zhaohan Ding, Siyuan Han, Wanrun Jiang, Guolin Ke, Zhaoqing Liu, Denghui Lu, Koki Muraoka, Hananeh Oliaei, Anurag Kumar Singh, Haohui Que, Weihong Xu, Zhangmancang Xu, Yong-Bin Zhuang, Jiayu Dai, Timothy J. Giese, Weile Jia, Ben Xu, Darrin M. York, Linfeng Zhang, Han Wang. “DeePMD-kit v3: A Multiple-Backend Framework for Machine Learning Potentials.” J. Chem. Theory Comput. 21 (2025): 4375-4385.

In addition, please follow this page to cite the methods you used.

Getting Started

• Getting Started
  • 1. Easy install
    • 1.1. Install DeePMD-kit in 1s
    • 1.2. Install off-line packages
    • 1.3. Install with conda
    • 1.4. Install with docker
    • 1.5. Install Python interface with pip
  • 2. DeePMD-kit Quick Start Tutorial
    • 2.1. Task
    • 2.2. Background
    • 2.3. Practice

Advanced

• 1. Backend
  • 1.1. Supported backends
    • 1.1.1. TensorFlow
    • 1.1.2. PyTorch
    • 1.1.3. JAX
    • 1.1.4. Paddle
    • 1.1.5. DP
  • 1.2. Switch the backend
    • 1.2.1. Training
    • 1.2.2. Inference
  • 1.3. Convert model files between backends
• 2. Installation
  • 2.1. Easy install
    • 2.1.1. Install DeePMD-kit in 1s
    • 2.1.2. Install off-line packages
    • 2.1.3. Install with conda
    • 2.1.4. Install with docker
    • 2.1.5. Install Python interface with pip
  • 2.2. Install from source code
    • 2.2.1. Install the Python interface
    • 2.2.2. Install the C++ interface
  • 2.3. Install from pre-compiled C library
    • 2.3.1. Use Pre-compiled C Library to build the LAMMPS plugin and i-PI driver
  • 2.4. Install LAMMPS
    • 2.4.1. Install LAMMPS’s DeePMD-kit module (built-in mode)
    • 2.4.2. Install LAMMPS (plugin mode)
  • 2.5. Install i-PI
  • 2.6. Building conda packages
  • 2.7. Install Node.js interface
    • 2.7.1. Install from npm
    • 2.7.2. Build from source
  • 2.8. Easy install the latest development version
    • 2.8.1. Install with docker
    • 2.8.2. Install with pip
    • 2.8.3. Download pre-compiled C Library
    • 2.8.4. Pre-release conda-forge packages
• 3. Data
  • 3.1. System
    • 3.1.1. Mixed type
  • 3.2. Formats of a system
    • 3.2.1. NumPy format
    • 3.2.2. HDF5 format
    • 3.2.3. Raw format and data conversion
  • 3.3. Prepare data with dpdata
• 4. Model
  • 4.1. Overall
    • 4.1.1. Theory
    • 4.1.2. Instructions
  • 4.2. Descriptor "se_e2_a"
    • 4.2.1. Theory
    • 4.2.2. Instructions
    • 4.2.3. Type embedding support
    • 4.2.4. Difference among different backends
    • 4.2.5. Model compression
  • 4.3. Descriptor "se_e2_r"
    • 4.3.1. Theory
    • 4.3.2. Instructions
    • 4.3.3. Type embedding support
    • 4.3.4. Difference among different backends
    • 4.3.5. Model compression
  • 4.4. Descriptor "se_e3"
    • 4.4.1. Theory
    • 4.4.2. Instructions
    • 4.4.3. Type embedding
    • 4.4.4. Difference among different backends
    • 4.4.5. Model compression
  • 4.5. Descriptor "se_atten"
    • 4.5.1. Theory
    • 4.5.2. Descriptor "se_atten"
    • 4.5.3. Type embedding
    • 4.5.4. Difference among different backends
    • 4.5.5. Type map
    • 4.5.6. Data format
    • 4.5.7. Model compression
    • 4.5.8. Training example
  • 4.6. Descriptor DPA-2
    • 4.6.1. Theory
    • 4.6.2. Requirements of installation
    • 4.6.3. Limitations of the JAX backend with LAMMPS
    • 4.6.4. Data format
    • 4.6.5. Type embedding
    • 4.6.6. Model compression
  • 4.7. Descriptor DPA3
    • 4.7.1. Theory
    • 4.7.2. Hyperparameter tests
    • 4.7.3. Requirements of installation from source code
    • 4.7.4. Limitations of the JAX backend with LAMMPS
    • 4.7.5. Data format
    • 4.7.6. Type embedding
    • 4.7.7. Model compression
  • 4.8. Descriptor DPA4
    • 4.8.1. Overview
    • 4.8.2. Model scaffold
    • 4.8.3. Descriptor construction
    • 4.8.4. Angular representation
    • 4.8.5. Radial basis and smooth cutoff
    • 4.8.6. Attention and focus streams
    • 4.8.7. Grid nonlinearities
    • 4.8.8. Environment-seeded initial features
    • 4.8.9. Zone bridging and ZBL
    • 4.8.10. Fitting network
    • 4.8.11. Configuration
    • 4.8.12. Training modes
    • 4.8.13. Spin
    • 4.8.14. Performance and hardware recommendations
    • 4.8.15. LoRA fine-tuning
    • 4.8.16. Export
    • 4.8.17. Data format
    • 4.8.18. Limitations
    • 4.8.19. Citation
  • 4.9. Descriptor "hybrid"
    • 4.9.1. Theory
    • 4.9.2. Instructions
    • 4.9.3. Type embedding
    • 4.9.4. Model compression
  • 4.10. Determine sel
  • 4.11. Fit energy
    • 4.11.1. Theory
    • 4.11.2. The fitting network
    • 4.11.3. Loss
  • 4.12. Fit spin energy
    • 4.12.1. Spin
    • 4.12.2. Spin Loss
    • 4.12.3. Data format
  • 4.13. Fit energy Hessian
    • 4.13.1. Energy Hessian Loss
    • 4.13.2. Hessian Format in PyTorch
    • 4.13.3. Train the Model
    • 4.13.4. Test the Model
  • 4.14. Fit tensor like Dipole and Polarizability
    • 4.14.1. Theory
    • 4.14.2. The fitting Network
    • 4.14.3. Loss
    • 4.14.4. Training Data Preparation
    • 4.14.5. Train the Model
    • 4.14.6. Difference among different backends
  • 4.15. Fit electronic density of states (DOS)
    • 4.15.1. The fitting Network
    • 4.15.2. Loss
    • 4.15.3. Training Data Preparation
    • 4.15.4. Train the Model
    • 4.15.5. Test the Model
  • 4.16. Fit other properties
    • 4.16.1. The fitting Network
    • 4.16.2. Loss
    • 4.16.3. Training Data Preparation
    • 4.16.4. Train the Model
    • 4.16.5. Test the Model
    • 4.16.6. Data Normalization
  • 4.17. Type embedding approach
    • 4.17.1. Theory
    • 4.17.2. Instructions for TensorFlow backend
    • 4.17.3. Instructions for other backends
  • 4.18. Descriptor "se_a_mask"
    • 4.18.1. Type embedding
    • 4.18.2. Model compression
  • 4.19. Descriptor "se_e3_tebd"
    • 4.19.1. Theory
    • 4.19.2. Instructions
    • 4.19.3. Type embedding
    • 4.19.4. Model compression
  • 4.20. Deep potential long-range (DPLR)
    • 4.20.1. Theory
    • 4.20.2. Train a deep Wannier model for Wannier centroids
    • 4.20.3. Train the DPLR model
    • 4.20.4. Molecular dynamics simulation with DPLR
  • 4.21. Deep Potential - Range Correction (DPRc)
    • 4.21.1. Theory
    • 4.21.2. Training data
    • 4.21.3. Training the DPRc model
    • 4.21.4. Run MD simulations
    • 4.21.5. Pairwise DPRc
  • 4.22. Linear model
  • 4.23. Interpolation or combination with a pairwise potential
    • 4.23.1. Theory
    • 4.23.2. Table file
    • 4.23.3. Interpolation with a short-range pairwise potential
    • 4.23.4. Combination with a pairwise potential
  • 4.24. Change the model output bias for trained model
  • 4.25. Floating-point precision of the model
  • 4.26. Use dp show to show the model information
    • 4.26.1. Command Syntax
    • 4.26.2. Example Usage
    • 4.26.3. Output Description
    • 4.26.4. Example Output
  • 4.27. Use dp pretrained to download built-in models
    • 4.27.1. Command syntax
    • 4.27.2. Available built-in models
    • 4.27.3. Examples
    • 4.27.4. Use downloaded models in DeepPot
• 5. Training
  • 5.1. Train a model
  • 5.2. Advanced options
    • 5.2.1. Learning rate
    • 5.2.2. Optimizer
    • 5.2.3. Training parameters
    • 5.2.4. Options and environment variables
    • 5.2.5. Adjust sel of a frozen model
  • 5.3. Learning rate
    • 5.3.1. Quick Start
    • 5.3.2. Common parameters
    • 5.3.3. Exponential Decay Schedule
    • 5.3.4. Cosine Annealing Schedule
    • 5.3.5. Warmup-Stable-Decay Schedule
    • 5.3.6. Warmup Mechanism
    • 5.3.7. Mathematical Theory
    • 5.3.8. Migration from versions before 3.1.3
    • 5.3.9. References
  • 5.4. Training Parameters
    • 5.4.1. Writing JSON files using Visual Studio Code
  • 5.5. Parallel training
    • 5.5.1. TensorFlow Implementation
    • 5.5.2. PyTorch Implementation
    • 5.5.3. Paddle Implementation
  • 5.6. Multi-task training
    • 5.6.1. Theory
    • 5.6.2. Perform the multi-task training using PyTorch
    • 5.6.3. Finetune from the pre-trained multi-task model
    • 5.6.4. Multi-task specific parameters
  • 5.7. TensorBoard Usage
    • 5.7.1. Highlighted features
    • 5.7.2. How to use Tensorboard with DeePMD-kit
    • 5.7.3. PyTorch Profiler With TensorBoard
    • 5.7.4. Examples
    • 5.7.5. Attention
  • 5.8. Known limitations of using GPUs
  • 5.9. Finetune the pre-trained model
    • 5.9.1. TensorFlow Implementation
    • 5.9.2. PyTorch Implementation
    • 5.9.3. Paddle Implementation
• 6. Freeze and Compress
  • 6.1. Freeze a model
  • 6.2. Compress a model
    • 6.2.1. Theory
    • 6.2.2. Instructions
    • 6.2.3. Requirements of installation
• 7. Test
  • 7.1. Test a model
    • 7.1.1. Evaluate descriptors
  • 7.2. Calculate Model Deviation
    • 7.2.1. Theory
    • 7.2.2. Instructions
    • 7.2.3. Relative model deviation
• 8. Inference
  • 8.1. Python interface
    • 8.1.1. External neighbor list algorithm
  • 8.2. C/C++ interface
    • 8.2.1. C++ interface
    • 8.2.2. C interface
    • 8.2.3. Header-only C++ library interface (recommended)
  • 8.3. Node.js interface
• 9. Command line interface
  • 9.1. Named Arguments
  • 9.2. Valid subcommands
  • 9.3. Sub-commands
    • 9.3.1. transfer
    • 9.3.2. train
    • 9.3.3. freeze
    • 9.3.4. test
    • 9.3.5. eval-desc
    • 9.3.6. compress
    • 9.3.7. doc-train-input
    • 9.3.8. model-devi
    • 9.3.9. convert-from
    • 9.3.10. neighbor-stat
    • 9.3.11. change-bias
    • 9.3.12. train-nvnmd
    • 9.3.13. gui
    • 9.3.14. convert-backend
    • 9.3.15. show
    • 9.3.16. pretrained
• 10. Integrate with third-party packages
  • 10.1. Use deep potential with dpdata
  • 10.2. Use deep potential with ASE
  • 10.3. Run MD with LAMMPS
    • 10.3.1. units
    • 10.3.2. Enable DeePMD-kit plugin (plugin mode)
    • 10.3.3. pair_style deepmd
    • 10.3.4. pair_style deepspin
    • 10.3.5. Compute tensorial properties
    • 10.3.6. Combine with other commands
  • 10.4. Run path-integral MD with i-PI
  • 10.5. Running MD with GROMACS
  • 10.6. Interfaces out of DeePMD-kit
    • 10.6.1. Plugins
    • 10.6.2. C/C++ interface used by other packages
    • 10.6.3. Command line interface used by other packages
• 11. Agent Skills
  • 11.1. List of skills
  • 11.2. Related reference
  • 11.3. Install skills
    • 11.3.1. If you are a user
    • 11.3.2. If you are an agent
  • 11.4. Minimal verification
• 12. Use NVNMD
• 13. Runtime environment variables
  • 13.1. All interfaces
  • 13.2. Environment variables of dependencies
  • 13.3. Python interface only
  • 13.4. C++ interface only
• 14. FAQs
  • 14.1. How to tune Fitting/embedding-net size ?
    • 14.1.1. Al2O3
    • 14.1.2. Cu
    • 14.1.3. Water
    • 14.1.4. Mg-Al
  • 14.2. How to control the parallelism of a job?
    • 14.2.1. MPI or multiprocessing (optional)
    • 14.2.2. Parallelism between independent operators
    • 14.2.3. Parallelism within individual operators
    • 14.2.4. Tune the performance
  • 14.3. Do we need to set rcut < half boxsize?
  • 14.4. How to set sel?
  • 14.5. Installation
    • 14.5.1. Inadequate versions of gcc/g++
    • 14.5.2. Build files left in DeePMD-kit
  • 14.6. The temperature undulates violently during the early stages of MD
  • 14.7. MD: cannot run LAMMPS after installing a new version of DeePMD-kit
  • 14.8. Model compatibility
  • 14.9. Why does a model have low precision?
    • 14.9.1. Data
    • 14.9.2. Model
    • 14.9.3. Training

Tutorial

• Tutorials
• Publications

Developer Guide

• Find DeePMD-kit C/C++ library from CMake
• Create a model in TensorFlow
  • Design a new component
  • Register new arguments
  • Package new codes
    • Package customized C++ OPs
• Create a model in other backends
  • Design a new component
    • New descriptors
    • New fitting nets
    • New models
    • Floating-point precision
  • Register new arguments
  • Package new codes
    • Package customized C++ OPs
  • Unit tests
    • Universal tests
    • Consistent tests
• Atom Type Embedding
  • Overview
  • Preliminary
  • How to use
  • Code Modification
    • trainer (train/trainer.py)
    • model (model/ener.py)
    • embedding net (descriptor/se*.py)
    • fitting net (fit/ener.py)
• Coding Conventions
  • Preface
  • Python
    • Rules
    • Whitespace
    • General advice
    • Writing documentation in the code
  • C++
  • Run scripts to check the code
• CI/CD
  • CI
    • Test CUDA
  • CD
• Python API
  • Submodules
  • Functions
  • Package Contents
    • DeepPotential()
• OP API
  • op_module
    • AddFltNvnmd()
    • Annotated
    • Any
    • ConvertForwardMap()
    • CopyFltNvnmd()
    • Descrpt()
    • DescrptNorot()
    • DescrptSeA()
    • DescrptSeAEf()
    • DescrptSeAEfPara()
    • DescrptSeAEfVert()
    • DescrptSeAMask()
    • DescrptSeR()
    • DotmulFltNvnmd()
    • DprcPairwiseIdx()
    • EwaldRecp()
    • FltNvnmd()
    • Gelu()
    • GeluCustom()
    • GeluGrad()
    • GeluGradCustom()
    • GeluGradGrad()
    • GeluGradGradCustom()
    • MapAparam()
    • MapFltNvnmd()
    • MatmulFitnetNvnmd()
    • MatmulFlt2fixNvnmd()
    • MatmulFltNvnmd()
    • MulFltNvnmd()
    • NeighborStat()
    • PairTab()
    • ParallelProdForceSeA()
    • ProdEnvMatA()
    • ProdEnvMatAMix()
    • ProdEnvMatAMixNvnmdQuantize()
    • ProdEnvMatANvnmdQuantize()
    • ProdEnvMatR()
    • ProdForce()
    • ProdForceNorot()
    • ProdForceSeA()
    • ProdForceSeAMask()
    • ProdForceSeR()
    • ProdVirial()
    • ProdVirialNorot()
    • ProdVirialSeA()
    • ProdVirialSeR()
    • QuantizeNvnmd()
    • SoftMinForce()
    • SoftMinSwitch()
    • SoftMinVirial()
    • TabulateFusion()
    • TabulateFusionGrad()
    • TabulateFusionGradGrad()
    • TabulateFusionSeA()
    • TabulateFusionSeAGrad()
    • TabulateFusionSeAGradGrad()
    • TabulateFusionSeAtten()
    • TabulateFusionSeAttenGrad()
    • TabulateFusionSeAttenGradGrad()
    • TabulateFusionSeR()
    • TabulateFusionSeRGrad()
    • TabulateFusionSeRGradGrad()
    • TabulateFusionSeT()
    • TabulateFusionSeTGrad()
    • TabulateFusionSeTGradGrad()
    • Tanh4FltNvnmd()
    • UnaggregatedDy2Dx()
    • UnaggregatedDy2DxS()
    • UnaggregatedDyDx()
    • UnaggregatedDyDxS()
    • add_flt_nvnmd()
    • convert_forward_map()
    • copy_flt_nvnmd()
    • descrpt()
    • descrpt_norot()
    • descrpt_se_a()
    • descrpt_se_a_ef()
    • descrpt_se_a_ef_para()
    • descrpt_se_a_ef_vert()
    • descrpt_se_a_mask()
    • descrpt_se_r()
    • dotmul_flt_nvnmd()
    • dprc_pairwise_idx()
    • ewald_recp()
    • flt_nvnmd()
    • gelu()
    • gelu_custom()
    • gelu_grad()
    • gelu_grad_custom()
    • gelu_grad_grad()
    • gelu_grad_grad_custom()
    • map_aparam()
    • map_flt_nvnmd()
    • matmul_fitnet_nvnmd()
    • matmul_flt2fix_nvnmd()
    • matmul_flt_nvnmd()
    • mul_flt_nvnmd()
    • neighbor_stat()
    • pair_tab()
    • parallel_prod_force_se_a()
    • prod_env_mat_a()
    • prod_env_mat_a_mix()
    • prod_env_mat_a_mix_nvnmd_quantize()
    • prod_env_mat_a_nvnmd_quantize()
    • prod_env_mat_r()
    • prod_force()
    • prod_force_norot()
    • prod_force_se_a()
    • prod_force_se_a_mask()
    • prod_force_se_r()
    • prod_virial()
    • prod_virial_norot()
    • prod_virial_se_a()
    • prod_virial_se_r()
    • quantize_nvnmd()
    • soft_min_force()
    • soft_min_switch()
    • soft_min_virial()
    • tabulate_fusion()
    • tabulate_fusion_grad()
    • tabulate_fusion_grad_grad()
    • tabulate_fusion_se_a()
    • tabulate_fusion_se_a_grad()
    • tabulate_fusion_se_a_grad_grad()
    • tabulate_fusion_se_atten()
    • tabulate_fusion_se_atten_grad()
    • tabulate_fusion_se_atten_grad_grad()
    • tabulate_fusion_se_r()
    • tabulate_fusion_se_r_grad()
    • tabulate_fusion_se_r_grad_grad()
    • tabulate_fusion_se_t()
    • tabulate_fusion_se_t_grad()
    • tabulate_fusion_se_t_grad_grad()
    • tanh4_flt_nvnmd()
    • unaggregated_dy2_dx()
    • unaggregated_dy2_dx_s()
    • unaggregated_dy_dx()
    • unaggregated_dy_dx_s()
  • op_grads_module
    • Annotated
    • Any
    • ProdForceGrad()
    • ProdForceSeAGrad()
    • ProdForceSeAMaskGrad()
    • ProdForceSeRGrad()
    • ProdVirialGrad()
    • ProdVirialSeAGrad()
    • ProdVirialSeRGrad()
    • SoftMinForceGrad()
    • SoftMinVirialGrad()
    • prod_force_grad()
    • prod_force_se_a_grad()
    • prod_force_se_a_mask_grad()
    • prod_force_se_r_grad()
    • prod_virial_grad()
    • prod_virial_se_a_grad()
    • prod_virial_se_r_grad()
    • soft_min_force_grad()
    • soft_min_virial_grad()
• C++ API
  • Class Hierarchy
  • File Hierarchy
  • Full API
    • Namespaces
      • Namespace tensorflow
    • Classes and Structs
      • Struct NeighborListData
        • Struct Documentation
      • Class AtomMap
        • Class Documentation
      • Class DeepBaseModelBackend
        • Inheritance Relationships
        • Class Documentation
      • Class DeepPot
        • Inheritance Relationships
        • Class Documentation
      • Class DeepPotJAX
        • Inheritance Relationships
        • Class Documentation
      • Class DeepPotPD
        • Inheritance Relationships
        • Class Documentation
      • Class DeepPotTF
        • Inheritance Relationships
        • Class Documentation
      • Class DeepSpinBackend
        • Inheritance Relationships
        • Class Documentation
      • Class DeepSpinPT
        • Inheritance Relationships
        • Class Documentation
      • Class DeepTensor
        • Class Documentation
      • Class DeepTensorPT
        • Inheritance Relationships
        • Class Documentation
      • Class DipoleChargeModifier
        • Class Documentation
      • Class DipoleChargeModifierTF
        • Inheritance Relationships
        • Class Documentation
    • Enums
    • Functions
      • Function deepmd::convert_pbtxt_to_pb
        • Function Documentation
      • Function deepmd::get_backend
        • Function Documentation
      • Function deepmd::load_op_library
        • Function Documentation
      • Function deepmd::name_prefix
        • Function Documentation
      • Function deepmd::read_file_to_string
        • Function Documentation
      • Template Function deepmd::select_by_type
        • Function Documentation
      • Template Function deepmd::select_map(typename std::vector<VT>::iterator, const typename std::vector<VT>::const_iterator, const std::vector<int>&, const int&, const int&, const int&, const int&)
        • Function Documentation
      • Template Function deepmd::select_map_inv(typename std::vector<VT>::iterator, const typename std::vector<VT>::const_iterator, const std::vector<int>&, const int&)
        • Function Documentation
      • Template Function deepmd::select_real_atoms_coord
        • Function Documentation
      • Template Function deepmd::session_get_scalar
        • Function Documentation
      • Template Function deepmd::session_input_tensors(std::vector<std::pair<std::string, tensorflow::Tensor>>&, const std::vector<VALUETYPE>&, const int&, const std::vector<int>&, const std::vector<VALUETYPE>&, const double&, const std::vector<VALUETYPE>&, const std::vector<VALUETYPE>&, const deepmd::AtomMap&, const std::string, const bool)
        • Function Documentation
      • Template Function deepmd::session_input_tensors_mixed_type
        • Function Documentation
    • Typedefs
      • Typedef deepmd::STRINGTYPE
        • Typedef Documentation
• C API
  • Class Hierarchy
  • File Hierarchy
  • Full API
    • Namespaces
      • Namespace deepmd::hpp
        • Classes
        • Functions
    • Classes and Structs
      • Struct InputNlist
        • Struct Documentation
      • Struct DP_DeepBaseModelDevi
        • Inheritance Relationships
        • Struct Documentation
      • Struct DP_DeepPotModelDevi
        • Inheritance Relationships
        • Struct Documentation
      • Struct DP_DeepSpinModelDevi
        • Inheritance Relationships
        • Struct Documentation
      • Struct DP_DipoleChargeModifier
        • Struct Documentation
      • Class DeepBaseModel
        • Inheritance Relationships
        • Class Documentation
      • Class DeepPot
        • Inheritance Relationships
        • Class Documentation
      • Class DeepSpin
        • Inheritance Relationships
        • Class Documentation
      • Class DeepTensor
        • Class Documentation
    • Functions
      • Specialized Template Function _DP_DeepPotCompute< double >
        • Function Documentation
      • Template Function _DP_DeepPotComputeMixedType
        • Function Documentation
      • Specialized Template Function _DP_DeepPotComputeMixedType< float >
        • Function Documentation
      • Specialized Template Function _DP_DeepPotComputeNList< double >
        • Function Documentation
      • Template Function _DP_DeepPotModelDeviCompute
        • Function Documentation
      • Specialized Template Function _DP_DeepPotModelDeviCompute< float >
        • Function Documentation
      • Specialized Template Function _DP_DeepPotModelDeviComputeNList< double >
        • Function Documentation
      • Template Function _DP_DeepSpinCompute
        • Function Documentation
      • Specialized Template Function _DP_DeepSpinCompute< float >
        • Function Documentation
      • Specialized Template Function _DP_DeepSpinComputeNList< double >
        • Function Documentation
      • Template Function _DP_DeepSpinModelDeviCompute
        • Function Documentation
      • Specialized Template Function _DP_DeepSpinModelDeviCompute< float >
        • Function Documentation
      • Specialized Template Function _DP_DeepSpinModelDeviComputeNList< double >
        • Function Documentation
      • Template Function _DP_DeepTensorCompute
        • Function Documentation
      • Specialized Template Function _DP_DeepTensorCompute< float >
        • Function Documentation
      • Specialized Template Function _DP_DeepTensorComputeNList< double >
        • Function Documentation
      • Template Function _DP_DeepTensorComputeTensor
        • Function Documentation
      • Specialized Template Function _DP_DeepTensorComputeTensor< float >
        • Function Documentation
      • Specialized Template Function _DP_DeepTensorComputeTensorNList< double >
        • Function Documentation
      • Template Function _DP_DipoleChargeModifierComputeNList
        • Function Documentation
      • Specialized Template Function _DP_DipoleChargeModifierComputeNList< float >
        • Function Documentation
      • Function _DP_Get_Energy_Pointer(double&, const int)
        • Function Documentation
      • Function deepmd::hpp::convert_pbtxt_to_pb
        • Function Documentation
      • Template Function deepmd::hpp::select_by_type
        • Function Documentation
      • Template Function deepmd::hpp::validate_charge_spin
        • Function Documentation
      • Function DP_DeepBaseModelCheckOK
        • Function Documentation
      • Function DP_DeepBaseModelDeviGetCutoff
        • Function Documentation
      • Function DP_DeepBaseModelDeviGetDimFParam
        • Function Documentation
      • Function DP_DeepBaseModelDeviGetNumbTypesSpin
        • Function Documentation
      • Function DP_DeepBaseModelDeviIsAParamNAll
        • Function Documentation
      • Function DP_DeepBaseModelGetDimAParam
        • Function Documentation
      • Function DP_DeepBaseModelGetNumbTypes
        • Function Documentation
      • Function DP_DeepBaseModelGetTypeMap
        • Function Documentation
      • Function DP_DeepBaseModelIsAParamNAll
        • Function Documentation
      • Function DP_DeepPotCompute
        • Function Documentation
      • Function DP_DeepPotCompute3
        • Function Documentation
      • Function DP_DeepPotComputef2
        • Function Documentation
      • Function DP_DeepPotComputeMixedType
        • Function Documentation
      • Function DP_DeepPotComputeNList
        • Function Documentation
      • Function DP_DeepPotComputeNList3
        • Function Documentation
      • Function DP_DeepPotComputeNListf2
        • Function Documentation
      • Function DP_DeepPotGetCutoff
        • Function Documentation
      • Function DP_DeepPotGetDimChgSpin
        • Function Documentation
      • Function DP_DeepPotGetNumbTypes
        • Function Documentation
      • Function DP_DeepPotGetTypeMap
        • Function Documentation
      • Function DP_DeepPotIsAParamNAll
        • Function Documentation
      • Function DP_DeepPotModelDeviCompute
        • Function Documentation
      • Function DP_DeepPotModelDeviCompute3
        • Function Documentation
      • Function DP_DeepPotModelDeviComputef2
        • Function Documentation
      • Function DP_DeepPotModelDeviComputeNList
        • Function Documentation
      • Function DP_DeepPotModelDeviComputeNList3
        • Function Documentation
      • Function DP_DeepPotModelDeviComputeNListf2
        • Function Documentation
      • Function DP_DeepPotModelDeviGetCutoff
        • Function Documentation
      • Function DP_DeepPotModelDeviGetDimChgSpin
        • Function Documentation
      • Function DP_DeepPotModelDeviGetNumbTypes
        • Function Documentation
      • Function DP_DeepPotModelDeviHasDefaultFParam
        • Function Documentation
      • Function DP_DeepSpinCheckOK
        • Function Documentation
      • Function DP_DeepSpinComputef2
        • Function Documentation
      • Function DP_DeepSpinComputeNListf2
        • Function Documentation
      • Function DP_DeepSpinGetDimAParam
        • Function Documentation
      • Function DP_DeepSpinGetNumbTypes
        • Function Documentation
      • Function DP_DeepSpinGetTypeMap
        • Function Documentation
      • Function DP_DeepSpinIsAParamNAll
        • Function Documentation
      • Function DP_DeepSpinModelDeviCompute2
        • Function Documentation
      • Function DP_DeepSpinModelDeviComputeNList2
        • Function Documentation
      • Function DP_DeepSpinModelDeviGetCutoff
        • Function Documentation
      • Function DP_DeepSpinModelDeviGetDimFParam
        • Function Documentation
      • Function DP_DeepSpinModelDeviGetNumbTypesSpin
        • Function Documentation
      • Function DP_DeepSpinModelDeviIsAParamNAll
        • Function Documentation
      • Function DP_DeepTensorCompute
        • Function Documentation
      • Function DP_DeepTensorComputeNList
        • Function Documentation
      • Function DP_DeepTensorComputeTensor
        • Function Documentation
      • Function DP_DeepTensorComputeTensorNList
        • Function Documentation
      • Function DP_DeepTensorGetCutoff
        • Function Documentation
      • Function DP_DeepTensorGetNumbTypes
        • Function Documentation
      • Function DP_DeepTensorGetSelTypes
        • Function Documentation
      • Function DP_DeleteChar
        • Function Documentation
      • Function DP_DeleteDeepBaseModelDevi
        • Function Documentation
      • Function DP_DeleteDeepPotModelDevi
        • Function Documentation
      • Function DP_DeleteDeepSpinModelDevi
        • Function Documentation
      • Function DP_DeleteDipoleChargeModifier
        • Function Documentation
      • Function DP_DipoleChargeModifierCheckOK
        • Function Documentation
      • Function DP_DipoleChargeModifierComputeNListf
        • Function Documentation
      • Function DP_DipoleChargeModifierGetNumbSelTypes
        • Function Documentation
      • Function DP_DipoleChargeModifierGetSelTypes
        • Function Documentation
      • Function DP_NewDeepPotModelDevi
        • Function Documentation
      • Function DP_NewDeepPotWithParam
        • Function Documentation
      • Function DP_NewDeepSpin
        • Function Documentation
      • Function DP_NewDeepSpinModelDeviWithParam
        • Function Documentation
      • Function DP_NewDeepTensor
        • Function Documentation
      • Function DP_NewDipoleChargeModifier
        • Function Documentation
      • Function DP_NewNlist
        • Function Documentation
      • Function DP_NlistCheckOK
        • Function Documentation
      • Function DP_NlistSetMask
        • Function Documentation
      • Function DP_ReadFileToChar
        • Function Documentation
      • Function DP_SelectByType
        • Function Documentation
    • Defines
      • Define DP_CHECK_OK
        • Define Documentation
      • Define DP_REQUIRES_OK
        • Define Documentation
    • Typedefs
      • Typedef DP_DeepBaseModelDevi
        • Typedef Documentation
      • Typedef DP_DeepPotModelDevi
        • Typedef Documentation
      • Typedef DP_DeepSpinModelDevi
        • Typedef Documentation
      • Typedef DP_DipoleChargeModifier
        • Typedef Documentation
• Core API
  • Class Hierarchy
  • File Hierarchy
  • Full API
    • Namespaces
      • Namespace std
    • Classes and Structs
      • Struct deepmd_exception_oom
        • Inheritance Relationships
        • Struct Documentation
      • Struct InputNlist
        • Struct Documentation
      • Template Struct DescrptSeRGPUExecuteFunctor
        • Struct Documentation
      • Template Struct GeluGradGPUExecuteFunctor
        • Struct Documentation
      • Template Struct ProdForceSeAGPUExecuteFunctor
        • Struct Documentation
      • Template Struct ProdVirialSeAGPUExecuteFunctor
        • Struct Documentation
      • Template Struct TabulateCheckerGPUExecuteFunctor
        • Struct Documentation
      • Template Struct TabulateFusionGradGPUExecuteFunctor
        • Struct Documentation
    • Unions
    • Functions
      • Function build_nlist(std::vector<std::vector<int>>&, std::vector<std::vector<int>>&, const std::vector<double>&, const int&, const double&, const double&, const std::vector<int>&, const std::vector<int>&, const std::vector<int>&, const std::vector<int>&, const SimulationRegion<double>&, const std::vector<int>&)
        • Function Documentation
      • Function build_nlist(std::vector<std::vector<int>>&, std::vector<std::vector<int>>&, const std::vector<double>&, const std::vector<int>&, const std::vector<int>&, const double&, const double&, const std::vector<int>&, const SimulationRegion<double>&)
        • Function Documentation
      • Function compute_descriptor(std::vector<double>&, std::vector<double>&, std::vector<double>&, const std::vector<double>&, const int&, const std::vector<int>&, const SimulationRegion<double>&, const bool&, const int&, const std::vector<int>&, const std::vector<int>&, const std::vector<int>&, const std::vector<int>&, const int, const int, const int, const int)
        • Function Documentation
      • Function compute_descriptor_se_a_ef_para
        • Function Documentation
      • Function compute_descriptor_se_a_extf
        • Function Documentation
      • Function compute_dRdT_1
        • Function Documentation
      • Function copy_coord
        • Function Documentation
      • Template Function deepmd::build_nlist_gpu
        • Function Documentation
      • Function deepmd::convert_nlist
        • Function Documentation
      • Template Function deepmd::convert_to_inter_cpu
        • Function Documentation
      • Template Function deepmd::convert_to_phys_cpu
        • Function Documentation
      • Template Function deepmd::copy_coord_cpu
        • Function Documentation
      • Function deepmd::cos_switch(const double&, const double&, const double&)
        • Function Documentation
      • Template Function deepmd::cprod
        • Function Documentation
      • Template Function deepmd::delete_device_memory
        • Function Documentation
      • Template Function deepmd::dot2
        • Function Documentation
      • Template Function deepmd::dot4
        • Function Documentation
      • Function deepmd::DPGetDeviceCount
        • Function Documentation
      • Function deepmd::DPSetDevice
        • Function Documentation
      • Template Function deepmd::env_mat_a_nvnmd_quantize_cpu
        • Function Documentation
      • Template Function deepmd::env_mat_r_cpu
        • Function Documentation
      • Function deepmd::filter_ftype_gpu
        • Function Documentation
      • Template Function deepmd::format_nlist_cpu
        • Function Documentation
      • Template Function deepmd::gelu_cpu
        • Function Documentation
      • Template Function deepmd::gelu_grad_cpu
        • Function Documentation
      • Template Function deepmd::gelu_grad_grad_cpu
        • Function Documentation
      • Function deepmd::group_atoms_cpu
        • Function Documentation
      • Template Function deepmd::invsqrt
        • Function Documentation
      • Specialized Template Function deepmd::invsqrt< float >
        • Function Documentation
      • Template Function deepmd::malloc_device_memory(FPTYPE *&, const int)
        • Function Documentation
      • Template Function deepmd::malloc_device_memory_sync(FPTYPE *&, const std::vector<FPTYPE>&)
        • Function Documentation
      • Template Function deepmd::malloc_device_memory_sync(FPTYPE *&, std::vector<FPTYPE>&)
        • Function Documentation
      • Function deepmd::max_numneigh
        • Function Documentation
      • Template Function deepmd::memcpy_device_to_host(const FPTYPE *, FPTYPE *, const int)
        • Function Documentation
      • Template Function deepmd::memcpy_host_to_device(FPTYPE *, const FPTYPE *, const int)
        • Function Documentation
      • Template Function deepmd::memset_device_memory
        • Function Documentation
      • Template Function deepmd::normalize_coord_cpu
        • Function Documentation
      • Template Function deepmd::pair_tab_cpu
        • Function Documentation
      • Template Function deepmd::prod_env_mat_a_gpu
        • Function Documentation
      • Template Function deepmd::prod_env_mat_r_cpu
        • Function Documentation
      • Template Function deepmd::prod_force_a_cpu(FPTYPE *, const FPTYPE *, const FPTYPE *, const int *, const int, const int, const int, const int)
        • Function Documentation
      • Template Function deepmd::prod_force_a_gpu
        • Function Documentation
      • Template Function deepmd::prod_force_grad_a_gpu
        • Function Documentation
      • Template Function deepmd::prod_force_grad_r_gpu
        • Function Documentation
      • Template Function deepmd::prod_force_r_gpu
        • Function Documentation
      • Template Function deepmd::prod_virial_a_gpu
        • Function Documentation
      • Template Function deepmd::prod_virial_grad_a_gpu
        • Function Documentation
      • Template Function deepmd::prod_virial_grad_r_gpu
        • Function Documentation
      • Template Function deepmd::prod_virial_r_gpu
        • Function Documentation
      • Template Function deepmd::soft_min_switch_force_cpu
        • Function Documentation
      • Template Function deepmd::soft_min_switch_virial_cpu
        • Function Documentation
      • Function deepmd::spline3_switch
        • Function Documentation
      • Template Function deepmd::tabulate_fusion_se_a_cpu
        • Function Documentation
      • Template Function deepmd::tabulate_fusion_se_a_grad_cpu
        • Function Documentation
      • Template Function deepmd::tabulate_fusion_se_a_grad_grad_cpu
        • Function Documentation
      • Template Function deepmd::tabulate_fusion_se_r_cpu
        • Function Documentation
      • Template Function deepmd::tabulate_fusion_se_r_grad_cpu
        • Function Documentation
      • Template Function deepmd::tabulate_fusion_se_r_grad_grad_cpu
        • Function Documentation
      • Template Function deepmd::tabulate_fusion_se_t_cpu
        • Function Documentation
      • Template Function deepmd::tabulate_fusion_se_t_grad_cpu
        • Function Documentation
      • Template Function deepmd::tabulate_fusion_se_t_grad_grad_cpu
        • Function Documentation
      • Template Function deepmd::tabulate_fusion_se_t_tebd_cpu
        • Function Documentation
      • Template Function deepmd::tabulate_fusion_se_t_tebd_grad_cpu
        • Function Documentation
      • Template Function deepmd::tabulate_fusion_se_t_tebd_grad_grad_cpu
        • Function Documentation
      • Template Function deepmd::test_encoding_decoding_nbor_info_gpu
        • Function Documentation
      • Function deepmd::use_nei_info_gpu
        • Function Documentation
      • Template Function deepmd::volume_cpu
        • Function Documentation
      • Template Function dotmul_flt_nvnmd
        • Function Documentation
      • Function DPAssert(hipError_t, const char *, int, bool)
        • Function Documentation
      • Function env_mat_r
        • Function Documentation
      • Template Function find_max_expo(int64_t&, T *, int64_t, int64_t)
        • Function Documentation
      • Function format_nlist_i_fill_a
        • Function Documentation
      • Function nborAssert(cudaError_t, const char *, int, bool)
        • Function Documentation
      • Function omp_get_num_threads
        • Function Documentation
      • Template Function split_flt
        • Function Documentation
    • Variables
    • Defines
      • Define DPErrcheck
        • Define Documentation
      • Define GPU_MAX_NBOR_SIZE
        • Define Documentation
      • Define gpuDeviceSynchronize
        • Define Documentation
      • Define gpuGetLastError
        • Define Documentation
      • Define gpuMemcpy
        • Define Documentation
      • Define gpuMemcpyDeviceToDevice
        • Define Documentation
      • Define gpuMemcpyDeviceToHost
        • Define Documentation
      • Define gpuMemcpyHostToDevice
        • Define Documentation
      • Define gpuMemset
        • Define Documentation
      • Define MOASPNDIM
        • Define Documentation
      • Define NBIT_FLTF
        • Define Documentation
      • Define nborErrcheck
        • Define Documentation
      • Define TPB
        • Define Documentation
    • Typedefs
      • Typedef uint_64
        • Typedef Documentation

Project Details

• License
• Authors and Credits
  • Cite DeePMD-kit and methods
  • Package Contributors
  • Other Credits
• Logo

• Index

• Module Index

• Search Page