search

twitter-research / graph-neural-pde

Graph Neural PDEs
Apache License 2.0
317 stars 52 forks source link

20 random initializations and 100 random train-val-test splits #7

Closed alip67 closed 2 years ago

alip67 commented 3 years ago

Hi,

could you explain how you generate the setting for table 2 in the paper?

in the paper, you state that " for all datasets using 100 random splits with 20 random initializations." which means that the set_train_val_test_split function should generate these splits. could you elaborate on how these parts work in your code?

melifluos commented 2 years ago

Yes, I appreciate that this is now very late, but I only just noticed the issue. It happens in ray_tune.py L154.

In our pipeline we use RayTune to do a hyperparameter search and then we evaluate the best hyperparameters by repeating 100 times with 20 random splits ie. 20 random seeds.