MLServer
MLServer copied to clipboard
Allow to load settings and model-settings from CLI flags
Currently, mlserver relies on having settings.json and model-settings.json files present or falling back to environment variables. It would be good to also allow users to specify these flags directly through the CLI.
For that, we should look for an integration between Pydantic (what we use to define the settings parameters) and some CLI library. We are currently using click for our CLI, but it doesn't seem that both projects are integrated.
https://github.com/facebookresearch/hydra might be interesting option to look at (for unified config file / CLI arguments handling)
Thanks for sharing @yonil7! Do you know if hydra integrates with Pydantic to expose CLI options?
I don't know about Pydantic integration. It just seems like a promising library to me that might take the pain from CLI args / config file args unification.