Zuobai Zhang

Yes, you're right. The problem is caused by `range`, while the behavior of `config_dict()` is correct. In this case, I would suggest to manually convert `range` into list when dumping...

Hi, this issue is better raised in the [GearNet](https://github.com/DeepGraphLearning/GearNet) repo. Typically, we only load one split (with at most 22,000 proteins) and use the round-robin method for pre-training. So will...

Thanks. I guess that maybe it is still too large for a split to fit in your memory. In this case, you may need a machine with large cpu memory....

Sorry for the inconvience. I've changed rdkit version to 2021.9.2, which can address the problem in my case. Maybe you can have a try. If it doesn't work for you,...

Hi, maybe you can try the code below. ```python ... from torchdrug import data pdb_file = utils.download("https://files.rcsb.org/download/1NSK.pdb", "./") p1 = data.Protein.from_pdb(pdb_file, atom_feature=None, bond_feature=None, residue_feature="default") pdb_file_2 = utils.download("https://files.rcsb.org/download/1EJ4.pdb", "./") p2 =...

Hi! I haven't found repetitive urls in the `datasets.AlphaFoldDB` class. Could you point out the repetitive urls you mentioned?

Hi, I don't think this should be a dataset-specific problem. It seems that you fail to build the `torch_extension` in TorchDrug. Could you check this?

Hi, please find the datasets here (https://zenodo.org/records/7593591), or you can use the code to download it automatically. Also, a processed version can be found in the CDConv repo (https://github.com/hehefan/Continuous-Discrete-Convolution).

Hi, could you please re-check the downloaed files? There should be a file called `nrPDB-EC_annot.tsv` in the `EnzymeCommission.zip` file, which contains the labels.

I think it's okay to use v4 instead of v2. The pre-training dataset doesn't have a large effect on the final performance.