You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In this work, we aim to aid the training of Hamiltonian-predicting models and NNPs for druglike
molecules by significantly extending and improving the nablaDFT dataset [11]. We double the number
of molecules and conformations to 1,936,929 and 12,676,264, respectively.
May I know while database file contains these 12,676,264 conformations with energy and forces?
The text was updated successfully, but these errors were encountered:
Hello! At this point forces are available for all data splits apart from dataset_train_full and dataset_test_conformations_full and stored in ase_db files.
All the databases are here: nablaDFT/links/energy_databases.json or available through the lightning pipeline.
The databases contain formation energy as well.
Alternately, separate conformations are available as an archive of xyz files here : archive and energies with other DFT properties are stored in csv file: csv file
May I know while database file contains these 12,676,264 conformations with energy and forces?
The text was updated successfully, but these errors were encountered: