stable-baselines3
Use this skill for reinforcement learning tasks including training RL agents (PPO, SAC, DQN, TD3, DDPG, A2C, etc.), creating custom Gym environments, implementing callbacks for monitoring and control, using vectorized environments for parallel training, and integrating with deep RL workflows. This skill should be used when users request RL algorithm implementation, agent training, environment design, or RL experimentation.
Security Assessment
Detected risks:
About stable-baselines3
The stable-baselines3 skill is designed for practitioners and researchers engaged in reinforcement learning (RL) tasks. It provides reliable implementations of various RL algorithms such as PPO, SAC, DQN, TD3, DDPG, and A2C, which are essential for training RL agents effectively. By leveraging this skill, users can train agents in a streamlined manner, create custom environments, and monitor agent performance through callbacks, addressing common challenges in RL experimentation and implementation.
FAQ
What algorithms are supported by stable-baselines3?
Stable Baselines3 supports several algorithms including PPO, SAC, DQN, TD3, DDPG, and A2C.
Can I use custom environments with stable-baselines3?
Yes, you can create custom environments that inherit from gymnasium.Env and implement the required methods.
What are the requirements for custom environments?
Custom environments must implement __init__(), reset(), step(), render(), and close() methods, and adhere to specific constraints regarding observation formats.
Does stable-baselines3 support parallel training?
Yes, it supports vectorized environments for parallel training using DummyVecEnv or SubprocVecEnv.
Are there limitations on action spaces in custom environments?
Yes, SB3 does not support Discrete or MultiDiscrete spaces with start!=0.
Install stable-baselines3
Quick Setup:
- Copy the skill folder to
.claude/skills/ - Claude will automatically detect and use the skill
Repository
davila7/claude-code-templates