📚 Documentation

The documentation about check pointing claims:

To change the checkpoint path use the default_root_dir argument:
# saves checkpoints to 'some/path/' at every epoch end
trainer = Trainer(default_root_dir="some/path/")

However, if a logger is configured, the checkpoints go magically to the path of the logger and not to default_root_dir:

trainer = L.Trainer(
  max_epochs=EPOCHS,
  logger=CSVLogger("save_dir", "exp_name"),
  default_root_dir="default_root_dir"
)
trainer.fit(...)

It's even more magic when you add several loggers (seems to take the path of the first logger)

I'm missing such thinks in the documentation...

cc @lantiga @Borda