Published March 25, 2026 | Version v1
Dataset Open

E3RelaxH2 dataset

Authors/Creators

Description

Five datasets used in E3RelaxH2:  `xmno`, `mp`, `c2db`, `vdW`, and `jarvis`.

Files

Files (100.7 MB)

Name Size Download all
md5:ab87581d115b57dd001ad5260ed35f2e
5.0 MB Download
md5:8f6016113b12f774b17e1d1bf225e36e
15.7 MB Download
md5:8e9c03422d1215b34f1d4322afadd03d
65.9 MB Download
md5:2c51917bea911752a2732ad1cccc2817
2.7 MB Download
md5:80d1c2444c2de28d222361ad7e30a4bd
11.4 MB Download

Additional details

Identifiers

URL
https://github.com/guaguabujianle/E3RelaxH2

Related works

Is described by
https://github.com/guaguabujianle/E3RelaxH2 (URL)

Dates

Submitted
2026-03-25

References

  • Yang Z, Zhao Y M, Wang X, et al. Equivariant Atomic and Lattice Modeling Using Geometric Deep Learning for Crystal Structure Optimization[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2026, 40(33): 27747-27755.