Supporting Files to Machine Learning-Based Prediction of the Glass Transition Temperature of Organic Compounds Using Experimental Data

This bundle consists of the main Python script, a readme.txt, a requirements_console_script.txt specifying the required packages, and the trained model modes as pickle les (pickle is a package for saving and loading trained algorithms). In the Python script, explanations for the correct input format are given along with examples. The code also enables multi-component input, which may be a fast and convenient option for some applications.

The following pickle files (https://docs.python.org/3/library/pickle.html) are available:

sm : SMILES Mode, requires SMILES string + melting temperature
sm_no_tm : SMILES Mode without Tm, requires SMILES string
fg_cho : Functional Group Mode for CHO compounds, requires functional groups + melting temperature
fg_cho_no_tm : Functional Group Mode for CHO compounds without Tm, requires functional groups
fg_nhal : Functional Group Mode for CHO compounds containing nitrogen or halogen atoms, requires functional groups + melting temperature
fg_nhal_no_tm : Functional Group Mode for CHO compounds containing nitrogen or halogen atoms, requires functional groups

For further information and contact please visit https://tgml.chemie.uni-bielefeld.de