pufferlib

SKILL.md

PufferLib is a high-performance reinforcement learning library designed for fast parallel environment simulation and training. It achieves training at millions of steps per second through optimized vectorization, native multi-agent support, and efficient PPO implementation (PuffeRL). The library provides the Ocean suite of 20+ environments and seamless integration with Gymnasium, PettingZoo, and specialized RL fra...

PuffeRL is PufferLib's optimized PPO+LSTM training algorithm achieving 1M-4M steps/second.

Use the template script: scripts/envtemplate.py provides complete single-agent and multi-agent environment templates with examples of:

Esta habilidad debe usarse cuando se trabaja con tareas de aprendizaje por refuerzo, incluido el entrenamiento de RL de alto rendimiento, el desarrollo de entornos personalizados, la simulación paralela vectorizada, los sistemas multiagente o la integración con entornos de RL existentes (Gymnasium, PettingZoo, Atari, Procgen, etc.). Utilice esta habilidad para implementar capacitación de PPO, crear entornos PufferEnv, optimizar el rendimiento de RL o desarrollar políticas con CNN/LSTM. Fuente: ovachiever/droid-tings.

Ver original