Implementation-MolGAN-PyTorch
Implementation-MolGAN-PyTorch copied to clipboard
dagouchin
PyTorch implementation of MolGAN: MolGAN: An implicit generative model for small molecular graphs.
Hello, Thank you for your code. But when I run main_gan and I set WGAN-Lambda = 0.0, the model can't generate a molecular graph, but some strange graphs. What's wrong...
I found the 'data_loader.py' is include in your code, but not used. Can you tell me how to use ‘data_loader.py’ in your model for data loading? Thanks.
Thank you for your exciting work! But when I was running the GAN model, I encountered a problem. ———————————————————————— Start training... [W python_anomaly_mode.cpp:104] Warning: Error detected in AddmmBackward. Traceback of...
I am facing this error while training the model after 21st epoch. RuntimeError: CUDA out of memory. Tried to allocate 2.00 MiB (GPU 0; 10.92 GiB total capacity; 9.98 GiB...
Hi Qin, Great work! Do you have the environment file of conda for running your code? Best, Ken
For how much epochs / image samples did you ran the model to get he results you added in your readme.md file for GAN model
https://github.com/kfzyqin/Implementation-MolGAN-PyTorch/blob/d07a490c841df7644011fe87c796978427fa828b/layers.py#L44 i.e., ```python3 for i in range(len(self.units)): in_units = list([x + in_features for x in self.units]) ``` I don't see how it is any different from ```python3 in_units = list([x...
Hello friends ,During traing, Why are some results nan? valid: 0.00, unique: 0.00, novel: nan, NP: nan, QED: nan, Solute: nan, SA: nan, diverse: nan, drugcand: 0.12
