Abstract
The new version of HydraGNN v4.0 provides additional core capabilities, such as:
Inclusion of multi-body atomistic cluster expansion MACE, polarizable atom interaction neural network PAINN, and equivariant principal neighborhood aggregation (PNAEq) among the message passing layers supported
-Inclusion of graph transformers to directly model long-range interactions between nodes that are distant in the graph topology
Integration of graph transformers with message passing layers by combining the graph embedding generated by the two mechanisms, which allows for an improved expressivity of the HydraGNN architecture
Improved re-implementation of multi-task learning (MTL) to allow its use for stabilized training across imbalanced, multi-source, multi-fidelity data
Introduction of multi-task parallelism, a newly proposed type of model parallelism specifically for MTL architectures, which allows to dispatch different output decoding heads to different GPU devices
Integration of multi-task parallelism with pre-existing distributed data parallelism to enable a 2D parallelization for distributed training
Improved portability of the distributed training across Intel GPUs, which has been testes on ALCF exascale supercomputer Aurora
Inclusion of 2-level fine-grained energy profilers portable across NVIDIA, AMD, and Intel GPUs to monitor the power and energy consumption associated with different functions executed by the HydraGNN code during data pre-load and training
Restructuring of previous examples and inclusion of new sets of examples to
More>>
- Developers:
-
.png "ORCID")
[1]
;
[1]
;
[1]
;
[1]
;
[1]
;
[1]
;
[2]
;
[2]
;
Baker, Justin
[3]
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
- Georgia Institute of Technology, Atlanta, GA (United States)
- University o California, Los Angeles, California, USA
- Release Date:
- 2025-10-16
- Project Type:
- Open Source, Publicly Available Repository
- Software Type:
- Scientific
- Programming Languages:
-
Python
- Version:
- v4.0
- Licenses:
-
BSD 3-clause "New" or "Revised" License
- Sponsoring Org.:
-
USDOE Office of Science (SC)Primary Award/Contract Number:AC05-00OR22725
- Code ID:
- 164024
- Research Org.:
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
- Country of Origin:
- United States
Citation Formats
Lupo Pasini, Massimiliano, Choi, Jong Youl, Mehta, Kshitij, Zhang, Pei, Weaver, Rylie, Chowdhury, Arindam, Li, Chaojian, Ye, Zhifan, and Baker, Justin.
HydraGNN v4.0.
Computer Software.
https://github.com/ORNL/HydraGNN/releases/tag/v4.0.
USDOE Office of Science (SC).
16 Oct. 2025.
Web.
doi:10.11578/dc.20250912.22.
Lupo Pasini, Massimiliano, Choi, Jong Youl, Mehta, Kshitij, Zhang, Pei, Weaver, Rylie, Chowdhury, Arindam, Li, Chaojian, Ye, Zhifan, & Baker, Justin.
(2025, October 16).
HydraGNN v4.0.
[Computer software].
https://github.com/ORNL/HydraGNN/releases/tag/v4.0.
https://doi.org/10.11578/dc.20250912.22.
Lupo Pasini, Massimiliano, Choi, Jong Youl, Mehta, Kshitij, Zhang, Pei, Weaver, Rylie, Chowdhury, Arindam, Li, Chaojian, Ye, Zhifan, and Baker, Justin.
"HydraGNN v4.0." Computer software.
October 16, 2025.
https://github.com/ORNL/HydraGNN/releases/tag/v4.0.
https://doi.org/10.11578/dc.20250912.22.
@misc{
doecode_164024,
title = {HydraGNN v4.0},
author = {Lupo Pasini, Massimiliano and Choi, Jong Youl and Mehta, Kshitij and Zhang, Pei and Weaver, Rylie and Chowdhury, Arindam and Li, Chaojian and Ye, Zhifan and Baker, Justin},
abstractNote = {The new version of HydraGNN v4.0 provides additional core capabilities, such as:
Inclusion of multi-body atomistic cluster expansion MACE, polarizable atom interaction neural network PAINN, and equivariant principal neighborhood aggregation (PNAEq) among the message passing layers supported
-Inclusion of graph transformers to directly model long-range interactions between nodes that are distant in the graph topology
Integration of graph transformers with message passing layers by combining the graph embedding generated by the two mechanisms, which allows for an improved expressivity of the HydraGNN architecture
Improved re-implementation of multi-task learning (MTL) to allow its use for stabilized training across imbalanced, multi-source, multi-fidelity data
Introduction of multi-task parallelism, a newly proposed type of model parallelism specifically for MTL architectures, which allows to dispatch different output decoding heads to different GPU devices
Integration of multi-task parallelism with pre-existing distributed data parallelism to enable a 2D parallelization for distributed training
Improved portability of the distributed training across Intel GPUs, which has been testes on ALCF exascale supercomputer Aurora
Inclusion of 2-level fine-grained energy profilers portable across NVIDIA, AMD, and Intel GPUs to monitor the power and energy consumption associated with different functions executed by the HydraGNN code during data pre-load and training
Restructuring of previous examples and inclusion of new sets of examples to illustrate the download, preprocess, and training of HydraGNN models on new large-scale open-source datasets for atomistic materials modeling (e.g., Alexandria, Transition1x, OMat24, OMol25)},
doi = {10.11578/dc.20250912.22},
url = {https://doi.org/10.11578/dc.20250912.22},
howpublished = {[Computer Software] \url{https://doi.org/10.11578/dc.20250912.22}},
year = {2025},
month = {oct}
}