Drug Molecule Generation with VAE (keras.io)
WGAN-GP with R-GCN for the generation of small molecular graphs (keras.io)
I try to generate novel compounds by using GAN and VAE with a portion of ZINC Dataset. However, when I follow the code from these two samples. They generated something that out of scope or not even looking decent molecule.
Thus, I have something to discuss with whoever see this one. Do you think that can we design a new drug by using another dataset and redefine GAN or VAE model?
- Badges
- Science method
More Tawatchai Jitporn's questions See All
Similar questions and discussions