---
license: mit
name: MP-20-CSP
tags:
  - MP-20
  - crystal-generation
  - materials-design
  - materials-discovery
  - structure-prediction
  
---
# Description
This is an [OMatG (Open Materials Generation)](https://github.com/FERMat-ML/OMatG) model for crystal structure prediction 
(CSP) of inorganic crystals trained on the MP-20 dataset.  
  
The MP-20 dataset consists of structures from the Materials Project database containing 20 or fewer atoms.
  
The subdirectories in this repository contain various model hyperparameters and training checkpoints for a 
variety of MP-20-CSP models.  

# Uses
The checkpoints and model hyperparameters can be used for prediction of crystalline structures with 
OMatG, 
as described in the [this README.md file](https://github.com/FERMat-ML/OMatG/blob/main/README.md)

# Recommendations
The [Linear-ODE checkpoints](https://huggingface.co/OMatG/MP-20-CSP/tree/main/Linear-ODE) 
currently provide the best results with 
respect to the match rate and the average root-mean square distance between generated and matched reference structures.

# Citation
Please cite [our paper on OpenReview](https://openreview.net/forum?id=gHGrzxFujU) if using OMatG.

# Links
[OMatG on GitHub](https://github.com/FERMat-ML/OMatG): See this repository for OMatG installation, training and usage instructions.  
[KIM Initiative](https://kim-initiative.org/): Knowledgebase of Interatomic Models. Tools and resources for researchers in materials science and chemistry.  
[Fermat-ML on GitHub](https://github.com/FERMat-ML): Foundational Representation of Materials. Machine learning foundation model for materials and chemistry discovery.