Transferability of Atom-Based Neural Networks Datasets
<p>This repository contains the datasets used in our article "Transferability of Atom-Based Neural Networks".</p> <p>The datasets are in the extxyz format, which can be used to train NequIP models. </p> <p>The extended xyz files contain all necessary information for the training of the models. The first four columns contain the xyz representation (Ångstrom), while the last column contains the atomic energies (kcal/mol), computed using either the AO-based EDA or MO-based IBO/IAO decomposition scheme. The total atomization energy (kcal/mol) is in the line below the number of atoms.</p> <p>The files are named as {dataset}_{decomp}.xyz.</p>
ShareScore
32/100
Overall dataset sharing score
Score breakdown
These five areas show where the dataset supports — or may limit — practical reuse.
- Stewardship
- 4
- Harmonization
- 4
- Access
- 16
- Reuse readiness
- 8
- Engagement
- 0
ScienceDex for faculty
Your academic CV and NIH biosketch
Make your CV in your school's format and your NIH biosketch from it, then download both as Word files.