search

RingBDStack / SUGAR

Code for "SUGAR: Subgraph Neural Network with Reinforcement Pooling and Self-Supervised Mutual Information Mechanism"
53 stars 6 forks source link

Exploding gradients problem and binary DD release #5

Closed id3509 closed 2 years ago

id3509 commented 2 years ago

1.Could you please give me the binary DD output and release again? The link is dead。 2.I also encountered the problem of exploding gradients, could you please give the details of the training parameters for the NCI1 or NCI109 dataset? Such as learning rate, batch size, hidden layer settings, etc. I shall be profoundly grateful if you will give me an answer.

Suchun-sv commented 2 years ago

Thanks for your attention.

  1. the binary DD: https://pan.baidu.com/s/1mkh7osIplj85QdhEtEdBCw 6m14

2.The parameters: we use the optuna framework (https://optuna.org/) for hyperparameter tuning on each dataset. The tuning range is 1e-3 to 1e-5 for learning rate, 32 for batch size, 128 for hidden dimension.