atomistic-machine-learning / schnetpack

SchNetPack - Deep Neural Networks for Atomistic Systems
Other
791 stars 215 forks source link

What can also be used as input modules in the AtomisticTask for energy prediction except PairwiseDistances? #565

Closed hekaido closed 1 year ago

hekaido commented 1 year ago

I'm trying to perform an energy prediction for structures from aflow dataset and I'm curious, if i can use more input modules than just spk.atomistic.PairwiseDistances Can i use all modules from spk.atomistic? If so, can you give me some examples of their usage?

jnsLs commented 1 year ago

The model architectures implemented in SchNetPack are designed to take pairwise distance (vectors) and atom types as inputs. The other modules in spk.atomistic are rather output modules.

hekaido commented 1 year ago

Thanks for your answer!