Performance/Configurations for ZINC dataset?

Hello, thank you for sharing this wonderful work!
I was wondering if you have tested this method on the ZINC dataset (e.g., ZINC250k)?  
 If so, could you clarify if the training and evaluation protocols are the same as those used for QM9, or if specific hyperparameters need to be adjusted?
Thanks!