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
As part of its multiple training schemes, it currently uses molfeat's MoleculeTransformer to generate ECFPs from SMILES and integrates the features directly in a Lightning module and associated training loop. The MoleculeTransformer can be passed with custom parameters to the Pytorch dataset, so with a little elbow grease it could potentially support most of the molfeat featurizers. I'd be willing to work out the integration if the project garnered any interest. I'll keep you fellas posted.
Back when I started you guys were very responsive, which gave me the confidence to integrate molfeat into my project! Thanks for the great work.
The text was updated successfully, but these errors were encountered:
Details
Hey fellas,
I made a project with Lightning + Hydra integration, to train models that capture Activity Cliff relationships.
https://github.com/cmvcordova/acsuite
As part of its multiple training schemes, it currently uses molfeat's
MoleculeTransformer
to generate ECFPs from SMILES and integrates the features directly in a Lightning module and associated training loop. The MoleculeTransformer can be passed with custom parameters to the Pytorch dataset, so with a little elbow grease it could potentially support most of the molfeat featurizers. I'd be willing to work out the integration if the project garnered any interest. I'll keep you fellas posted.Back when I started you guys were very responsive, which gave me the confidence to integrate molfeat into my project! Thanks for the great work.
The text was updated successfully, but these errors were encountered: