Stable Baselines

E99657

Python library reinforcement learning library software library

Stable Baselines is a popular Python library that provides reliable, well-tested implementations of reinforcement learning algorithms built on top of OpenAI Baselines.

Jump to: Surface forms Statements Referenced by

Observed surface forms (2)

Surface form	Occurrences
Stable Baselines3	2
Stable-Baselines3 (via wrappers)	1

Statements (50)

Predicate	Object
instanceOf	Python library ⓘ reinforcement learning library ⓘ software library ⓘ
basedOn	OpenAI Baselines ⓘ
category	machine learning software ⓘ open-source software ⓘ
compatibleWith	NumPy ⓘ TensorFlow (original versions) ⓘ
developedIn	Python ecosystem ⓘ
focusesOn	ease of use for RL practitioners ⓘ reproducible reinforcement learning research ⓘ
goal	provide reliable RL algorithm implementations ⓘ standardize RL research codebases ⓘ
hasDocumentation	online documentation website ⓘ
hasFeature	logging utilities ⓘ model saving and loading ⓘ tensorboard integration ⓘ unified interface for RL algorithms ⓘ vectorized environments ⓘ
hasSuccessor	Stable Baselines self-linksurface differs ⓘ surface form: Stable Baselines3
hostedOn	GitHub ⓘ
implements	reinforcement learning algorithms ⓘ
license	MIT License NERFINISHED ⓘ
programmingLanguage	Python ⓘ
provides	reliable implementations of RL algorithms ⓘ well-tested implementations of RL algorithms ⓘ
relatedTo	OpenAI Baselines ⓘ Stable Baselines self-linksurface differs ⓘ surface form: Stable Baselines3
supports	actor-critic methods ⓘ continuous action spaces ⓘ discrete action spaces ⓘ parallel environment execution ⓘ policy gradient methods ⓘ value-based methods ⓘ
supportsAlgorithm	A2C ⓘ ACKTR ⓘ DDPG ⓘ DQN ⓘ PPO ⓘ SAC ⓘ TD3 ⓘ
supportsEnvironmentInterface	Gymnasium ⓘ OpenAI Gym ⓘ
targetUser	data scientists ⓘ machine learning researchers ⓘ reinforcement learning practitioners ⓘ
usedFor	applied reinforcement learning projects ⓘ benchmarking RL algorithms ⓘ training reinforcement learning agents ⓘ
writtenIn	Python ⓘ

Referenced by (5)

Full triples — surface form annotated when it differs from this entity's canonical label.

PettingZoo → compatibleWith → Stable Baselines ⓘ

this entity surface form: Stable-Baselines3 (via wrappers)

Stable Baselines → hasSuccessor → Stable Baselines self-linksurface differs ⓘ

this entity surface form: Stable Baselines3

OpenAI Baselines → inspired → Stable Baselines ⓘ

OpenAI Baselines → relatedTo → Stable Baselines ⓘ

Stable Baselines → relatedTo → Stable Baselines self-linksurface differs ⓘ

this entity surface form: Stable Baselines3