Minima-Preserving Neural Network (MPNN) is a small Python library which builds an approximation of a potential energy surface given while respecting local minima locations and values provided by the user.
SupramolecularVAE is an open-source multi-component variational autoencoder for the property-guided inverse design of reticular frameworks including metal-organic frameworks and covalent-organic frameworks. Example datasets and Python notebooks are provided.
SorbMetaML is an open‐source meta-learning model for the prediction of unary adsorption for nanoporous materials based on example adsorption data for a material. SorbMetaML has been used to identify the optimal hydrogen storage temperature with the highest working capacity for a given pressure difference for diverse nanoporous materials. Datasets for the hydrogen adsorption of all-silica zeolites, hyper-cross-linked polymers, and metal-organic frameworks are provided.
SorbNet is a free, open-source deep neural network for the prediction of adsorption data for binary mixtures over large temperature and pressure ranges that can be used to optimize adsorption/desorption conditions. Example datasets and Python notebooks are provided.
DeePMD-kit is an open-source package for constructing deep-neural network (DNN) representations of the interatomic potential energy surface (PES) derived from ab-initio data. The DNN-PES reproduces the ab-initio PES with high accuracy and is extensive and differentiable. DeePMD-kit is interfaced with the popular open-source tool LAMMPS to perform large-scale molecular dynamics (MD) simulations.
Chemical Variational Autoencoder (chemical_VAE) is a free, open-source software for machine learning of molecular properties. chemical_VAE utilizes molecular SMILES that are encoded into a code vector representation and can be decoded from the code representation back to molecular SMILES. The autoencoder may also be jointly trained with property prediction to help shape the latent space. The new latent space can then be optimized upon to find the molecules with the most optimized properties of interest.