Resuming training from checkpoint results in "No train_dataloader() method defined" error #17693
Unanswered
geomlyd
asked this question in
Lightning Trainer API: Trainer, LightningModule, LightningDataModule
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I'm trying to resume training a model from a checkpoint, which, as far as I understand from the docs, should have all the information necessary for the Trainer to resume training (see Restoring Training State).
However, when calling trainer.fit with exactly those two arguments, I receive the following error
Exception has occurred: MisconfigurationException (note: full exception trace is shown but execution is paused at: _run_module_as_main) No `train_dataloader()` method defined. Lightning `Trainer` expects as minimum a `training_step()`, `train_dataloader()` and `configure_optimizers()` to be defined.
Isn't this supposed to restore the data module as well, at least according to this section of the docs? Am I doing something wrong?
Beta Was this translation helpful? Give feedback.
All reactions