Huge chunk of the dataset is invalid
#1
by
Tehemton
- opened
There are missing brackets in a huge chunk of the dataset which leads to them not translating into valid molecules. Curious to know if the authors have a way of mitigating that? Iterating over the dataset to delete those rows is cumbersome due to the size of the dataset. Any advice here is welcome.
Also wanted to understand if that error is coming from the SAFE encoding algorithm or something else.