atomistic.DipoleMoment

class atomistic.DipoleMoment(*args: Any, **kwargs: Any)

Predicts dipole moments from latent partial charges and (optionally) local, atomic dipoles. The latter requires a representation supplying (equivariant) vector features.

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

[2]

Gastegger, Behler, Marquetand. Machine learning molecular dynamics for the simulation of infrared spectra. Chemical science 8.10 (2017): 6924-6935.

[3]

Veit et al. Predicting molecular dipole moments by combining atomic partial charges and atomic dipoles. The Journal of Chemical Physics 153.2 (2020): 024113.

Parameters:

• n_in – input dimension of representation

• n_hidden – size of hidden layers. If an integer, same number of node is used for all hidden layers resulting in a rectangular network. If None, the number of neurons is divided by two after each layer starting n_in resulting in a pyramidal network.

• n_layers – number of layers.

• activation – activation function

• predict_magnitude – If true, calculate magnitude of dipole

• return_charges – If true, return latent partial charges

• dipole_key – the key under which the dipoles will be stored

• charges_key – the key under which partial charges will be stored

• correct_charges – If true, forces the sum of partial charges to be the total charge, if provided, and zero otherwise.

• use_vector_representation – If true, use vector representation to predict local, atomic dipoles.