This folder contains the code used for ML experiments. This is the code for running ML models and it is based on a clone of the ML code in the tmQMg ML code repo.
In addition to the NBO-graph based Gilmer models in the tmQMg ML repo this folder contains our fingerprint and SMILES graph based models.
There are 3 main driver scripts. See the argparser in each script for possible input parameters.
Fingerprint models:
ml_fingerprint.py
SMILES graph models:
ml_smiles_graph.py
To run the NBO-graph based Gilmer model from tmQMg use
ml_nbo.py