search

choderalab / espaloma

Extensible Surrogate Potential of Ab initio Learned and Optimized by Message-passing Algorithm 🍹https://arxiv.org/abs/2010.01196
https://docs.espaloma.org/en/latest/
MIT License
202 stars 23 forks source link

Mmff94 legacy typing with rdkit #203

Closed pmorerio closed 4 months ago

pmorerio commented 6 months ago

Hi, I added this feature.

I have verified that the experiment you show in fig. 2 of the paper reaches 98% accuracy on ZINC with this typing. (I had an initial bug related to atom ordering which is solved by using the openff method .to_rdkit() which give a perfect matching)

Hope it can be useuful.

Best,

P.

ijpulidos commented 4 months ago

Hello, thanks for your interest in our model and contributing to it. Could you be more explicit as to what is the motivation for these changes?