The number of tokens is inconsistent with the tokens provided in `LogP_optimization_demo.ipynb`, and cannot be learned using migration? about release HOT 4 OPEN

What kinds of extra characters do you have? You probably need to standardize your SMILEs (remove metals, mixtures, stereochemistry, etc.).

from release.

@isayev
My SMILES contains extra characters a, because the characters contain Na, Ca, what do you mean by standardized SMILES? What do I need to do? Thank you

from release.

hi @zhouhao-learning ,
Did you solve your problem? I am facing the same issue. I you have the solution please enlighten me.

from release.

hi @zhouhao-learning ,
Did you solve your problem? I am facing the same issue. I you have the solution please enlighten me.

Although the question is old, I'm answering it now because it seems it still unresolved...

Basically, the point is that you generally don't want ions (Na+, Ca2+) in your compound library, since they are just counterions to your compound. So, you need to remove those from the your SMILES data before using it.

Take a look at: https://molvs.readthedocs.io/en/latest/

Best.

from release.