You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
You can load atom descriptors from a sdf, npz or pkl file via the command line option --atom_descriptors_path <path_to_file> and use the descriptors either via --atom_descriptors feature (as extra atom features during the molecule featurization, that is, they are added to the regular atom features before the MPNN) or via --atom_descriptors descriptor (atomic descriptors are concatenated to the atomic feature vector after MPNN before aggregating the atoms to a molecule).
The data file should look like:
.npz descriptors are saved as 2D array for each molecule in the order of that in the data.csv
.pkl / .pckl / .pickle containing a pandas dataframe with smiles as index and numpy array of descriptors as columns
.sdf containing all mol blocks with descriptors as entries
How can I add atom_descriptors to your model?
The text was updated successfully, but these errors were encountered: