ai_research

Reduces code & configuration boilerplate with:

• Hydra for task configuration.
• Hydra-zen for Hydra structured configuration management.
• Lightning for PyTorch code.

Simplifies machine learning workflows:

• SLURM job definition, queuing and monitoring with Submitit through its Hydra Launcher plugin.
• Podman environment containerization for both regular & SLURM-based execution.
• Transition from regular execution to SLURM-based execution by swapping as little as a single Hydra configuration field.

Facilitates experimentation development through:

• An object-oriented common/project/task structure.
• Deeply integrated logging with Weights & Biases.

Promotes high-quality and reproducible code by:

• Formatting with Black.
• DType & Shape type hinting for PyTorch tensors using jaxtyping.
• Dynamic type-checking w/ Beartype.