ORNL/HydraGNN
Distributed PyTorch implementation of multi-headed graph convolutional neural networks
This tool helps materials scientists and researchers efficiently predict properties of materials at scale. It takes in structural data of materials, represented as graphs, and outputs predicted material properties at both the atomic (node) and full material (graph) levels. The primary users are researchers in materials science and chemistry who work with complex material structures and require high-throughput property prediction, especially on supercomputing infrastructures.
103 stars. Available on PyPI.
Use this if you need to perform scalable, multi-headed predictions of material properties based on their atomic structures, leveraging distributed computing environments.
Not ideal if you are not working with graph-structured data in materials science or if you only require predictions for small datasets without distributed computing needs.
Stars
103
Forks
40
Language
Python
License
BSD-3-Clause
Category
Last pushed
Mar 12, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ORNL/HydraGNN"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
pyg-team/pytorch_geometric
Graph Neural Network Library for PyTorch
a-r-j/graphein
Protein Graph Library
raamana/graynet
Subject-wise networks from structural MRI, both vertex- and voxel-wise features (thickness, GM...
pykale/pykale
Knowledge-Aware machine LEarning (KALE): accessible machine learning from multiple sources for...
dmlc/dgl
Python package built to ease deep learning on graph, on top of existing DL frameworks.