trainer.fit()

[Libardo1]

Writing forecasting models in GluonTS with PyTorch

trainer = pl.Trainer(max_epochs=10, gpus=-1 if torch.cuda.is_available() else None)
trainer.fit(net, train_dataloader=data_loader)

error::
INFO:pytorch_lightning.utilities.rank_zero:GPU available: True, used: True
INFO:pytorch_lightning.utilities.rank_zero:TPU available: False, using: 0 TPU cores
INFO:pytorch_lightning.utilities.rank_zero:IPU available: False, using: 0 IPUs
INFO:pytorch_lightning.utilities.rank_zero:HPU available: False, using: 0 HPUs

TypeError Traceback (most recent call last)
in ()
1 trainer = pl.Trainer(max_epochs=10, gpus=-1 if torch.cuda.is_available() else None)
----> 2 trainer.fit(net, train_dataloader=data_loader)

TypeError: fit() got an unexpected keyword argument 'train_dataloader'

[jaheba]

It should be train_dataloaders with an s at the end: https://pytorch-lightning.readthedocs.io/en/stable/common/trainer.html#fit

train_dataloader was removed here: Lightning-AI/pytorch-lightning#10325