training-pipeline

What problem does it solve?

This Skill orchestrates RL training tasks by routing to specialized sub-skills and managing experiment campaigns.

Core Features & Use Cases

• Curriculum-learning routing: delegates to curriculum-learning skill to progressively introduce tasks.
• Hyperparameter optimization coordination: triggers and coordinates hyperparameter search with AutoML workflows.
• Reward/penalty engineering integration: coordinates adjustments across reward tuning components.
• Training campaign management: oversees multi-run experiments and result tracking.

Quick Start

Use the training-pipeline skill to initiate an automl-guided training campaign and route to curriculum-learning, hyperparameter-optimization, or reward-penalty-engineering as needed.