Ready to use Machine learning models to predict a property. : comp_chem


I have a dataset of currently roughly 1500 reaction barriers (one reaction partner is constant) and growing. I thought it might be fun to see if we can predict barrier heights from the structure of the changing reactant.

Anyone knows any papers that include their code where we could input or molecules in whatever format (SMILES or whatever) and a single value and train a model?

This is not for serious research, we would get in contact with experts for this. We just want to play a bit with the data.

Read more here: Source link