Rohit Gupta

update on PR: all that's been discussed [here](https://github.com/PyTorchLightning/pytorch-lightning/pull/11089#issuecomment-1005548992) is now added. Only the following proposal is missing and I need more time to think about how to go about this....

> IMO a callback is not the right fit here. this will override the behavior of fit/validate/test if used this way. why doesn't this use the loops API instead? I...

any suggestions for the test here? I am thinking of extending these with profilers since they cover all the cases with hooks. https://github.com/Lightning-AI/lightning/blob/master/tests/tests_pytorch/models/test_hooks.py

It was totally unintentional. The hook was missing from the logging `FxValidator` that enabled logging in this hook. The real intention was just to ensure these hooks are profiled, so...

Noob q: Is it possible that `num_nodes < world_size/gpus_per_device`, where the user has access to more nodes in their cluster env than they actually need for their experiment?

got it! haven't trained on multi-node, still learning!! :)

@carmocca mind review as code-owner?

@JustinGoheen added a commit to resolve the issue

agreed!

I don't think it's a bug since `LightningDataModule` doesn't have `batch_size` as an argument.