Removing this redundancy code (a remnant from the conversion to sdf files ) in main.py as it makes execution fail for SMILES string that cannot be standardized and results in a Null Output predictions for those instances. I then ran predictions with the previously used datasets, on the updated repo and predictions was gotten for all molecules.
Removing this redundancy code (a remnant from the conversion to sdf files ) in main.py as it makes execution fail for SMILES string that cannot be standardized and results in a Null Output predictions for those instances. I then ran predictions with the previously used datasets, on the updated repo and predictions was gotten for all molecules.
Results without code: eos30gr_pred.csv
Results with code: eos30gr_output.csv