Retrosynthesis with attention-based NMT model and chemical analysis of “wrong” predictions

We consider retrosynthesis to be a machine translation problem. Accordingly, we apply an attention-based and completely data-driven model named Tensor2Tensor to a data set comprising approximately 50 000 diverse reactions extracted from the United States patent literature. The model significantly ou...

Descripción completa

Detalles Bibliográficos
Autores principales: Duan, Hongliang, Wang, Ling, Zhang, Chengyun, Guo, Lin, Li, Jianjun
Formato: Online Artículo Texto
Lenguaje:English
Publicado: The Royal Society of Chemistry 2020
Materias:
Chemistry
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9047528/
https://www.ncbi.nlm.nih.gov/pubmed/35494683
http://dx.doi.org/10.1039/c9ra08535a

Internet

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9047528/
https://www.ncbi.nlm.nih.gov/pubmed/35494683
http://dx.doi.org/10.1039/c9ra08535a

Ejemplares similares