Moad dataset broken archive

gcorso / DiffDock

Implementation of DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking

https://arxiv.org/abs/2210.01776

MIT License

976 stars 238 forks source link

Moad dataset broken archive #198

Open rpowalski opened 3 months ago

rpowalski commented 3 months ago

Hello Team, I have some trouble with following the dataset loader for the Moad dataset. This line suggests that there should be a directory pdb_ligand in the moad archive.

I downloaded and extracted the archive shared in here, but this directory is missing. To be sure I did it twice on different machines.

Here is what I see when I extracted the archive. Can you please share some advice on this?

Alain-chavanieu commented 3 months ago

I had the same as the download failed twice. Then it was ok . the un tar take time. Howevar, I had another problem when I tried to use NotADirectoryError: [Errno 20] Not a directory: '.....DiffDock/data/BindingMOAD_2020_processed/pdb_protein/._6hd6_1_protein.pdb/._6hd6_1_protein.pdb_protein.pdb'

amorehead commented 3 months ago

@rpowalski, I also am not seeing the pdb_ligand directory in the downloaded MOAD archive. This is after trying to download it using wget three separate times.

Harper-Hua commented 2 months ago

Same here. Couldn't find pdb_ligand.

amorehead commented 2 months ago

@jsilter, any chance you might be able to confirm the issue we are encountering here?