Do you like the fact that Hydra (Facebook: AI Research) automatically adds command line arguments from your config file? (no need to write a single line of argparse)
Do you hate the fact that they don't let you return the config file: https://github.com/facebookresearch/hydra/issues/407
I like lightweight, I don't like (more like hate) packages that force me into a certain development pattern which is why I don't care much for Hydra or pytorch-lightning.
If you like me, just want to load config from a yaml file and also automatically overrride it from the command line when required, try out my above python package. Any feedback is appreciated too. Thanks!
Show HN: Lightweight YAML Config CLI for Deep Learning Projects | Heykuki News