Lightning-AI/pytorch-lightning

Distributed mode overwrites the user's choice for dataloaders shuffling

Open

#10 170 ouverte le 27 oct. 2021

Voir sur GitHub
 (8 commentaires) (0 réactions) (0 assignés)Python (3 233 forks)batch import
data handlingfeaturehelp wanted

Métriques du dépôt

Stars
 (26 687 stars)
Métriques de merge PR
 (Merge moyen 9j 15h) (3 PRs mergées en 30 j)

Description

🐛 Bug

I ordered my training data in a specific manner and passed it to the DataLoader with shuffle=False (I use reload_dataloaders_every_n_epochs=1 to control it every epoch). Then, I found out that running in distributed mode re-creates the train dataloader and always set shuffle=True. The opposite happens for the val dataloader.

I think that the re-creation of the dataloaders must be using the same shuffle setting the original dataloaders had. Is there any reason not to do so?


Here is the related code with the hard-coded shuffle values:

train - shuffle forced to be true https://github.com/PyTorchLightning/pytorch-lightning/blob/c33df2639f19d49c5e7520294e3221efe402d684/pytorch_lightning/trainer/data_loading.py#L330-L333

val - shuffle forced to be true https://github.com/PyTorchLightning/pytorch-lightning/blob/c33df2639f19d49c5e7520294e3221efe402d684/pytorch_lightning/trainer/data_loading.py#L442-L443

cc @borda @justusschock @awaelchli

Guide contributeur