Third-Party Tutorials¶
This page contains tutorials which are not maintained by Farama Foundation and, as such, cannot be guaranteed to function as intended.
If you’d like to contribute a tutorial, please reach out on Discord.
Third-Party tutorials about gymnasium¶
Johnny Code from Youtube¶
Watch this set of videos to train Gymnasium Reinforcement Learning environments using Q-Learning, Deep Q-Learning, and other algorithms. Environments include FrozenLake-v1, MountainCar-v0, CartPole-v1, and MuJoCo Humanoid-v4. The youtuber uses Python with VSCode as the development environment. Libraries include Farama Gymnasium, PyTorch, Stable Baselines3. All codes are made available on github.com/johnnycode8 repository.
freeCodeCamp from Youtube¶
This video tells the basics of reinforcement learning and how to implement it using Gymnasium.
Hugging Face Deep Reinforcement Learning Course¶
A free course that goes from the basics of RL to deep RL algorithms (Q-learning, DQN, policy gradients, PPO and more), with hands-on notebooks that train agents on Gymnasium environments such as LunarLander-v3 and FrozenLake-v1.
Third-Party tutorials about libraries that use gymnasium¶
AgileRL¶
AgileRL focuses on reducing the time taken for training models and hyperparameter optimisation (HPO) providing tutorials for using it with PPO, TD3 and Rainbow.
CleanRL¶
CleanRL provides single-file implementations of deep RL algorithms (PPO, DQN, SAC, TD3 and more) trained on Gymnasium environments, with documentation explaining each implementation detail and benchmark results.
RLlib¶
RLlib is Ray’s scalable RL library. Its environments guide shows how to train on registered or custom Gymnasium environments and scale data collection across many CPUs and GPUs.
Stable-Baselines3¶
Stable-Baselines3 provides reliable PyTorch implementations of common RL algorithms. Its examples page shows how to train, save and evaluate agents on Gymnasium environments, including vectorized environments and custom wrappers.