|
|
--- |
|
|
license: apache-2.0 |
|
|
datasets: |
|
|
- nimashoghi/mptrj |
|
|
pipeline_tag: graph-ml |
|
|
tags: |
|
|
- chemistry |
|
|
- diffusers |
|
|
- anemoi |
|
|
- medvae |
|
|
--- |
|
|
# Position-based Equivariant Graph Neural Network (`pos-egnn`) |
|
|
This repository contains PyTorch model for loading and performing inference using the `pos-egnn`, a foundation model for Chemistry and Materials. |
|
|
|
|
|
**GitHub**: https://github.com/IBM/materials/tree/main/models/pos_egnn |
|
|
|
|
|
**HuggingFace**: https://huggingface.co/ibm-research/materials.pos-egnn |
|
|
|
|
|
<p align="center"> |
|
|
<img src="posegnn.svg"> |
|
|
</p> |
|
|
|
|
|
## Introduction |
|
|
We present `pos-egnn`, a Position-based Equivariant Graph Neural Network foundation model for Chemistry and Materials. The model was pre-trained on 1.4M samples (i.e., 90%) from the Materials Project Trajectory (MPtrj) dataset to predict energies, forces and stress. `pos-egnn` can be used as a machine-learning potential, as a feature extractor, or can be fine-tuned for specific downstream tasks. |
|
|
|
|
|
Besides the model weigths `pos-egnn.v1-6M.pt` (download from [HuggingFace](https://huggingface.co/ibm-research/materials.pos-egnn)), we also provide an `example.ipynb` notebook (download from [GitHub](https://github.com/ibm/materials)), which demonstrates how to perform inference, feature extraction and molecular dynamics simulation with the model. |
|
|
|
|
|
For more information, please reach out to [email protected] and/or [email protected] |
