insilicomedicine / GENTRL

Generative Tensorial Reinforcement Learning (GENTRL) model
611 stars 218 forks source link

question about the sampling.py #33

Open SylviaSWY opened 3 years ago

SylviaSWY commented 3 years ago

Dear author, I have successfully used the code you provided in the example folder, and I have successfully run the train_rl.py and successfully saved the trained model, however I encountered problems when I continue to run the sampling.py, the question is : using the saved gentrl model after rl, the model could sampling just one kind of SMILES! As the picture showed below., what's the problem? Can you kindly help me to solve this problem?

微信图片_20210930034313 微信图片_20210930034304