hydra-configs-how-to-run-deep-learning-pipeline-with-hydra-2

hydra-configs-how-to-run-deep-learning-pipeline-with-hydra-2.py

from hydra import compose, initialize_config_dir
from omegaconf import OmegaConf
from src.train import train


def main() -> None:
    with initialize_config_dir(version_base="1.3", config_dir="."):
        cfg = compose(
            config_name="train.yaml", return_hydra_config=True, overrides=[]
        )
    print(OmegaConf.to_yaml(cfg))
    train(cfg)


if __name__ == "__main__":
    main()