This project aims to create a machine learning model that can predict the bioavailability of molecules. Several types of models are tested, including a feed-forward neural network, a graph neural network, a random forest model, an XGBoost model and a features neural network.
All our code can be found in the Code folder.
The data files are in the Data folder.
The required packages are listed in requirements.txt
- Bioavailibility:
- Aqueous solubility:
- Nina Belusa: https://github.com/NinaB99
- Jonathan Below: https://github.com/jcbelow
- Matthias Galka: https://github.com/MGalka66
This project was carried out as part of EPFL's "AI for Chemistry" course.