XinhaoLi74 / SmilesPE

SMILES Pair Encoding: A data-driven substructure representation of chemicals
https://xinhaoli74.github.io/SmilesPE/
Apache License 2.0
177 stars 30 forks source link

Distribution of SPE file #1

Open jannisborn opened 4 years ago

jannisborn commented 4 years ago

We are working towards including your package as a dependency into pytoda, our package for pytorch datasets that has various functionalities to deal with molecules and proteins. See the PR here.

We noticed that your code is Apache licensed but that the file with the SPE substructures is not part of the package (this file).

Do you think it would be possible for you to include the file as a package dependency? This would be ideal since if you update the file and bump the pypi version the file would also be updated in all packages that use SmilesPE as a dependency. It's super easy, see here how to do it.

If you instead prefer the current setting, this would create headache for all dependent packages since every new release of SmilesPE would require to check manually whether you updated the SPE_ChEMBL.txt. If you still prefer this, we might redistribute the file in pytoda directly. For this case, could you please let us know how we can reference you? We were planning to add a note in our README.md that we took this file from your repo.