representation.PaiNN

class representation.PaiNN(*args: Any, **kwargs: Any)

PaiNN - polarizable interaction neural network

References:

[1]

Schütt, Unke, Gastegger: Equivariant message passing for the prediction of tensorial properties and molecular spectra. ICML 2021, http://proceedings.mlr.press/v139/schutt21a.html

Parameters:

• n_atom_basis – number of features to describe atomic environments. This determines the size of each embedding vector; i.e. embeddings_dim.

• n_interactions – number of interaction blocks.

• radial_basis – layer for expanding interatomic distances in a basis set

• cutoff_fn – cutoff function

• activation – activation function

• shared_interactions – if True, share the weights across interaction blocks.

• shared_interactions – if True, share the weights across filter-generating networks.

• epsilon – numerical stability parameter

• nuclear_embedding – custom nuclear embedding (e.g. spk.nn.embeddings.NuclearEmbedding)

• electronic_embeddings – list of electronic embeddings. E.g. for spin and charge (see spk.nn.embeddings.ElectronicEmbedding)