Hello,
I am trying to use hydra to configure my deep learning repo. One thing that is unclear to me is how to debug modules which utilize hydra configuration. From the DVC documentation it looks like we are supposed to remove the @hydra.main
from our module, but then how can we access our config when we are not using dvc exp run
, e.g. when we want to debug the script? I tried to use dvc.api.params_show()
but it does not seem trigger hydra composition and writing of a params.yaml file.
What is the intended workflow here?
Thanks