IRAS ML Predictions

Contained in this repository are the large files required for running the code in the following GitHub repository: https://github.com/zfried/IRAS_ML_Predictions fullDataset.csv contains all of the SMILES strings and length 70 feature vectors of the molecules used to train the mol2vec model. mol2vec_model_final_70.pkl is the trained mol2vec model that produces length 70 molecular feature vectors. prediction_molecules.csv contains all of the SMILES strings and length 89 feature vectors (including isotopic encoding) of the 84,863 molecules that are inputted into the trained model.