There is a newer version of this record available.

Dataset Open Access

Retrained Models and Scripts for Aluminum at 298K and 933K

Fiedler, Lenz; Cangi, Attila

Retrained Models and Scripts for Aluminum at 298K and 933K

Authors

- Fiedler, Lenz (HZDR/CASUS)
- Cangi, Attila (HZDR/CASUS)

Affiliations:

HZDR - Helmholtz-Zentrum Dresden-Rossendorf
CASUS - Center for Advanced Systems Understanding

Data set description

This data sets contains models, scripts and inference results for aluminum at room temperature and the melting point. Training data, hyperparameters and general methodology follow Ref. [1]. The models here are retrained versions of the ones discussed in this publication, and therefore retrained versions of the models contained in Ref. [2]. As such, data from Ref. [2] has been used. Only a subset of models contained in Ref. [1] have been retrained, namely the room temperature model, one liquid and one solid melting point model with four training snapshot each, and the final melting point hybrid model (six training snapshots per phase). Furthermore, for both the hybrid melting temperature model and the room temperature model, multiple models with different initializations were trained.

All models were trained with the MALA code [3] version 1.2.1. They show better accuracy than their original counterparts, as they were trained using the inter-snapshot shuffling algorithm first discussed for the MALA code in Ref. [4].

[1] - "Accelerating finite-temperature Kohn-Sham density functional theory with deep neural networks", Physical Review B, doi.org/10.1103/PhysRevB.104.035120
[2] - "RODARE", doi.org/10.14278/rodare.2485 (v1.0.0)
[3] - "MALA", Zenodo, doi.org/10.5281/zenodo.5557254
[4] - "Machine learning the electronic structure of matter across temperatures", Physical Review B, doi.org/10.1103/PhysRevB.108.125146

Contents

- The models themselves, labeled as either Al298K or Al933K, given as one .zip file per model
    - For 933K, additionally "liquid", "solid" and "hybrid" denotes the training data set
    - For ensembles, a running index denotes the number in the ensemble
- Inference results, given as a single .zip file
    - For all models, band energy and total free energy results are given in the .csv format
        - The columns in these files correspond to "Calculated via DFT LDOS", "Calculated via ML-DFT LDOS", "Calculated via Kohn-Sham system", respectively
    - For some models, additionally the predicted electronic density and density of states on select snapshots is given
- Shuffling, training and testing scripts, given as a single .zip file
    - Scripts are ready-to-use with suitable MALA installation, however, correct data paths have to be filled in
    
   

Files (2.1 GB)
Name Size
Al_298K_0.zip
md5:fa0c7db7acc6fd01c978195428acd524
6.5 MB Download
Al_298K_1.zip
md5:5fdc227be9da7aa018817c354086df7f
6.5 MB Download
Al_298K_2.zip
md5:5b05860806c083e27246e2e1860afccd
6.5 MB Download
Al_298K_3.zip
md5:085624ab0156fb11b4de4f1825aa4dec
6.5 MB Download
Al_298K_4.zip
md5:e7d6619fb20b6a6f55e85e693fb56110
6.5 MB Download
Al_933K_hybrid_0.zip
md5:6b6bb7174c45d8ce0094e5416b3c2da3
133.8 MB Download
Al_933K_hybrid_1.zip
md5:5f2312e1aee22a9f38f4b07d094d0131
124.4 MB Download
Al_933K_hybrid_2.zip
md5:1406a724ce02553771a215c09b191489
124.4 MB Download
Al_933K_hybrid_3.zip
md5:cdbc14c98f8e70966b9edf4791fc7bbb
124.3 MB Download
Al_933K_hybrid_4.zip
md5:11551c0b75e1604c07e925cfb5d5acb9
124.4 MB Download
Al_933K_liquid.zip
md5:2f1f035b115834ec8c92460477414ef0
133.8 MB Download
Al_933K_solid.zip
md5:fcfe8510e8bec329d4b95385886919b2
133.8 MB Download
inference_results.zip
md5:e8046fc5b8433a8e04d1813fde4cc514
1.2 GB Download
scripts.zip
md5:5d9add1c61037d6efcdf86dfb1712717
4.8 kB Download
186
207
views
downloads
All versions This version
Views 18677
Downloads 20778
Data volume 29.4 GB11.3 GB
Unique views 14268
Unique downloads 8439

Share

Cite as