Molecular conditional generation and property analysis of non-fullerene acceptors with deep learning
The molecular conditional generation and property prediction models are built with Pytorch (>=v1.7) and DGL-LifeSci.
utils.py : Dataset preparation and utils function.
model.py : Generative and prediction model.
config.py : Parameters of the two models.
cgen.py : Code for training and testing the generative model.
pre.py : Code for training and testing the prediction model.
sample.py : Code for sampling.
[1] Peng, S.-P.; Zhao, Y. Convolutional Neural Networks for the Design and Analysis of Non-Fullerene Acceptors. J. Chem. Inf. Model. 2019, 59, 4993–5001. [Paper] [Code]
[1] Gehring, J.; Auli, M.; Grangier, D.; Yarats, D.; Dauphin, Y. N. Convolutional Sequence to Sequence Learning. 2017. [Paper] [Code]
[2] Bai, S.; Kolter, J. Z.; Koltun, V. An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling. 2018. [Paper] [Code]
[3] Veličković, P.; Cucurull, G.; Casanova, A.; Romero, A.; Liò, P.; Bengio, Y. Graph Attention Networks. 2018. [Paper]
[4] Lopez S A, Sanchez-Lengeling B, de Goes Soares J, et al. Design principles and top non-fullerene acceptor candidates for organic photovoltaics[J]. Joule, 2017, 1(4): 857-870. [Paper] [Code]