openff.nagl

A toolkit for the generation of neural network models for predicting molecule properties.

Modules

features

Atom and bond features for GNN models.

molecule

Structures for managing molecule graphs

nn

Components for constructing and processing GNN models

storage

Classes to handle storage of labelled data for training

toolkits

Extensions of the OpenFF Toolkit ToolkitWrappers for use with NAGL.

utils

Assorted utilities, including resonance structure generation

Classes

GNNModel

A model that applies a graph convolutional step followed by pooling and readout steps.