model_cards/regression_transformer_description.md

Concurrent sequence regression and generation for molecular language modeling

The Regression Transformer is a multitask Transformer that reformulates regression as a conditional sequence modeling task. This yields a dichotomous language model that seamlessly integrates property prediction with property-driven conditional generation. For details see the arXiv preprint, the development code and the GT4SD endpoint for inference.

Each algorithm_version refers to one trained model. Each model can be used for two tasks, either to predict one (or multiple) properties of a molecule or to generate a molecule (given a seed molecule and a property constraint).

For examples and documentation of the model parameters, please see below. Moreover, we provide a model card (Mitchell et al. (2019)) at the bottom of this page.