ml-validate

To validate Hydra configuration and PyTorch Lightning setup, run an automated script that checks YAML syntax, verifies config composition, and tests _target_ component instantiation to catch ML project readiness issues early.

ML training runs often fail due to broken Hydra configuration, invalid _target_ imports, missing project structure, or code quality issues. Running a pre-training validation script identifies these dependency and configuration failures before you execute experiments.

Check GPU availability and dependencies for a PyTorch Lightning project by running automated training readiness checks that verify CUDA availability, import required and optional dependencies, and execute a fast dev training run.

The best way to verify ML project structure is running a validation script that confirms required directories and key files like src/train.py and configs/config.yaml exist before you start your training experiments.

Yes, you can test model and DataModule instantiation without a full experiment by running a fast dev training loop that exercises Hydra _target_ paths and verifies your trainer components execute correctly.