andreamad8/todcl

Continual Learning for Task-Oriented Dialogue Systems

todcl's Issues

error in trainer.fit in pytorch-lightning

I have the following error in the running the command:
CUDA_VISIBLE_DEVICES=0 python train.py --task_type DST --CL EWC --reg 0.01 --n_epochs 1

File "train.py", line 221, in
train(hyperparams)
File "train.py", line 121, in train
trainer.fit(model, task_loader, val_loader[task_id])
..
..
..

File "/usr/local/lib/python3.7/dist-packages/pytorch_lightning/utilities/apply_func.py", line 99, in apply_to_collection
return function(data, *args, **kwargs)
File "/usr/local/lib/python3.7/dist-packages/pytorch_lightning/core/lightning.py", line 512, in __check_allowed
raise ValueError(f"self.log({name}, {value}) was called, but {type(v).__name__} values cannot be logged")
ValueError: self.log(val_loss, loss) was called, but str values cannot be logged raise ValueError(f"self.log({name}, {value}) was called, but {type(v).__name__} values cannot be logged")
ValueError: self.log(val_loss, loss) was called, but str values cannot be logged