Tai Bui

Got the same issue with MoleculeDataset ``` mol_dataset = torchdrug_data.dataset.MoleculeDataset() mol_dataset.load_csv('data_dippr_liquid_heat_capacity_smiles.csv', smiles_field='SMILES') model = models.RGCN(input_dim=mol_dataset.node_feature_dim, num_relation=mol_dataset.num_bond_type, hidden_dims=[10, 10, 10, 10]) task = tasks.GCPNGeneration(model, mol_dataset.atom_types, max_edge_unroll=12, max_node=38, criterion='nll') optimizer = optim.Adam(task.parameters(),...

@dannyl1u From the latest update, I can confirm that when I deleted a certain message, the subsequent messages were not deleted altogether. Therefore, it would not be an issue. Adding...

I confirmed this happened with safari on Mac. Changed to Edge (also on Mac) it works fine.

Hi @mattwthompson, Thanks for getting back to me. Please find below some example systems where they should be constructed in a specific way, hopefully you find this helpful. There are...

Hi @YuuuXie, In the Section 4 of the tutorial (mentioned above by @potus28), how do we do the optimisation for hyper-parameters after the offline training step with large values of...

HI Yu Xie, the first one works. Thanks

Thanks Martin. It's good to hear from you here. I will create a filter for this to select only suggestions that satisfy the outcome constraints. Agreed that it's not optimal...