These are preliminary reports that have not been peer-reviewed. They should not be regarded as conclusive, guide clinical practice/health-related behavior, or be reported in news media as established information. For more information, please see our FAQs.
A Graph-Convolutional Neural Network for Addressing Small-Scale Reaction Prediction
preprintsubmitted on 31.01.2021, 03:36 and posted on 02.02.2021, 06:27 by Yejian Wu, Chengyun Zhang, Ling Wang, Hongliang Duan
We describe a graph-convolutional neural network (GCN) model whose reaction prediction capable as potent as the transformer model on sufficient data, and adopt the Baeyer-Villiger oxidation to explore their performance differences on limited data. The top-1 accuracy of GCN model (90.4%) is higher than that of transformer model (58.4%).