num_workers > 0

[pavelbatyr]

Hi, and thanks for the code!

I suggest using num_workers > 0 in the dataloaders. It can speed up training significantly. Using 4 workers makes training 30% faster in my case.

[harryjo97]

Hi,

Thank you for the suggestion! We would add the num_workers feature to the following line.

GDSS/utils/data_loader.py

Line 12 in 4d96334

train_dl = DataLoader(train_ds, batch_size=config.data.batch_size, shuffle=True)

[pavelbatyr]

Thanks! There's also a dataloader for molecules in utils/data_loader_mol.py