materials-discovery.zip

This contains files and data needed to reproduce the machine learning results of the paper "Materials discovery using generative machine learning". Further information can be found at: https://github.com/newalexander/generative-materials-discovery
The data folder contains:
alignn/Ed_tern.csv: The materials project (MP) IDs and decomposition enthalpies of material structures. The decomposition enthalpies were obtained from the work of Bartel et al. (2020).pgcgm/Ed_tern.csv: The predicted decomposition enthalpies of the structures generated by the PGCGM (Zhao et al., 2023).pgcgm/cifs.zip: The CIFs for the structured generated by the PGCGM.The Ed_tern folder contains the trained ALIGNN (Choudhary et al., 2021) model we used to evaluate on PGCGM data.
References
Bartel, C. J., Trewartha, A., Wang, Q., Dunn, A., Jain, A., and Ceder, G. A critical examination of com- pound stability predictions from machine-learned for- mation energies. npj Computational Materials, 6(1): 97, Jul 2020. ISSN 2057-3960. doi: 10.1038/ s41524-020-00362-y. URL https://doi.org/10. 1038/s41524-020-00362-yChoudhary, K. and DeCost, B. Atomistic line graph neural network for improved materials property pre- dictions. npj Computational Materials, 7(1):185, Nov 2021. ISSN 2057-3960. doi: 10.1038/ s41524-021-00650-1. URL https://doi.org/10. 1038/s41524-021-00650-1Zhao, Y., Siriwardane, E. M. D., Wu, Z., Fu, N., Al- Fahdi, M., Hu, M., and Hu, J. Physics guided deep learning for generative design of crystal materials with symmetry constraints. npj Computational Materials, 9 (1):38, Mar 2023. ISSN 2057-3960. doi: 10.1038/ s41524-023-00987-9. URL https://doi.org/10. 1038/s41524-023-00987-9