Mirrored by Aurigene AI

Discovery stage: Lead optimization

ChemBERTa-2 (77M) pretrained with multitask regression over 200 normalised RDKit descriptors. A strong, cheap baseline for MoleculeNet-style ADMET endpoints.

Upstream: DeepChem/ChemBERTa-77M-MTR - all credit to the original authors; the model card and licence below are theirs.

Explore the rest of the catalogue: Molecule Explorer - Protein Target Explorer - Drug Discovery Model Hub

ChemBERTa-77M-MTR

See the upstream model card for full details, training data and citation.

Downloads last month: 10

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Space using Aurigene-AI/ChemBERTa-77M-MTR 1

Collection including Aurigene-AI/ChemBERTa-77M-MTR

Molecular Representation & Property Prediction

Collection

Transformer encoders for SMILES/graph/image molecular representations — embeddings, ADMET & MoleculeNet property prediction. • 3 items • Updated Sep 10