Extending the Lightning class (pytorch_adapt/frameworks/lightning/lightning.py)

[just-eoghan]

Suggested Feature

The Lightning class could be extended to include two more functions

1. test_step
2. test_epoch_end

Implementation

The test_step and test_epoch end could operate in the same way as the validation_step and epoch_end and return a test_score.

Reasoning

This would allow lightning users to specify a test_dataloader containing a hold-out set in their datamodule. The best saved model on validation data can then be run against the test data using pytorch lightning trainer.test function call.

[KevinMusgrave]

Makes sense