Project Overview¶
torch-molecule
is a package under active development to support molecular discovery using deep learning. It provides a simple, sklearn
-style interface and model checkpoints for fast deployment and benchmarking.
Main components:
Predictive Models - ✔ GREA, SGIR, IRM, GIN/GCN w/ virtual, DIR - ✔ SMILES-based LSTM/Transformers - ⏳ More models
Generative Models - ✔ Graph DiT, GraphGA, DiGress, MolGPT - ⏳ GDSS and more
Representation Models - ✔ MoAMa, AttrMasking, ContextPred, EdgePred, InfoGraph - ⏳ more models and pretrained checkpoints
Note
This project is in active development. Interfaces and features may change.
Project Structure¶
torch_molecule
├── base
├── encoder
├── generator
├── nn
├── predictor
└── utils
Acknowledgements¶
This project was adapted from python-project-template.