Separate *_percent_check for each *_dataloader

[rohitgr7]

🚀 Feature

Can we have *_percent_check to be a list too where len(*_percent_check) == len(*_dataloaders)? In case if it is int then it will be same for all the dataloaders passed. Don't know how this can be useful in any case, just a thought.

Motivation

Pitch

For each val_dataloader or test_dataloader we can have an option to pass *_percent_check as a list with a percent_check for each of the dataloader. For eg. val_percent_check = [0.1, 0.4] and val_dataloaders = [val_dl1, val_dl2].

Alternatives

Additional context

Later we can do the same for training as well if #1959 get's merged.

[awaelchli]

Your proposal is definitely feasible, but I'm not sure what the use case for this would be. Currently the *_percent_check args are intented to be used for debugging purposes, e.g. trying to overfit to make sure the model has the capacity, in which case a separation into multiple sizes is not needed.

Could be a follow up to your existing PR #1920
@PyTorchLightning/core-contributors

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[rohitgr7]

looks like not a useful feature.