ml-config-system

What problem does it solve?

This Skill provides a standardized, hierarchical, and framework-agnostic way to define and manage complex machine learning experiment configurations, reducing boilerplate and ensuring consistency.

Core Features & Use Cases

• Hierarchical Configuration: Define base configurations and inherit/extend them for specific tasks (SL Neural, SL Tree, RL).
• Built-in Infrastructure: Automatically includes output directory management, console logging, checkpointing, metrics logging, and MLflow tracking.
• Framework Agnostic: The specification defines semantics, allowing implementations to adapt to PyTorch, TensorFlow, Hugging Face, XGBoost, LightGBM, CatBoost, JAX, etc.
• Use Case: When setting up configurations for a new ML project, whether it's a deep learning model for image classification or a gradient boosting model for time-series forecasting, this Skill ensures all necessary components like logging, checkpointing, and parameter tracking are pre-configured.

Quick Start

Use the ml-config-system skill to generate a base configuration for an SL Neural Regression experiment.