This repository contains PySIDT, a package containing a set of low-data machine-learning algorithms for prediction of chemical properties based on the subgraph isomorphic tree generation (SIDT) approach originally developed in Johnson and Green 2024. While the algorithm in that work was specific to rate coefficients, this implementation can be applied to prediction of arbitrary properties. This implementation also incorporates uncertainty prepruning, as detailed in Pang et al. 2024.
- Install PySIDT from source
git clone https://github.com/zadorlab/PySIDT.gitcd PySIDTconda env create -f environment.ymlconda activate pysidt_envpip install -e .
- Install molecule from source
git clone https://github.com/ReactionMechanismGenerator/molecule.gitcd moleculeconda activate pysidt_envmakepip install -e .