I want to run the kge test command on the train split instead of the test split as a sanity check for my user-defined model (if the model is really learning anything meaningful) But I cannot really find a way to do this. Can you point me to how to specify the config arguments?
Hi,
I want to run the kge test command on the train split instead of the test split as a sanity check for my user-defined model (if the model is really learning anything meaningful) But I cannot really find a way to do this. Can you point me to how to specify the config arguments?
The one I tried is
And it does not work.