Using PiNN

Architecture

Components of PiNN are written in the language of Keras Layers and Models, those are referred to as PiNN layers and networks respectively.

PiNN layers are reusable operations in training ANNs, e.g. calculation of neighboring lists, or radial basis functions.

PiNN networks are defined ANN architectures that makes atomic predictions. Since PiNN networks are essentially Keras Models, they are ready for simple regression tasks.

Models

In addition to layers and networks, PiNN implement several models. PiNN models interpret the output of ANNs as physical quantities, e.g. atomic energies. Those models enables the training of quantities derives from the atomic predictions, like forces and dipole moments.

PiNN models are implemented as TensorFlow estimators. models are also responsible for interfacing with external libraries like ASE to run simulations.

What to read

Checkout quick start to get started. See the notebook examples for more examples.

If you are interested in a specific application, e.g. fitting a machine-learned potential, read the available options of the potential model and network (likely PiNet) you'd like to use.