Project
Agent RL LunarLander
Projet de reinforcement learning autour de LunarLander-v3 : entraînement d'un agent PPO avec Gymnasium et Stable-Baselines3, suivi d'expériences, API FastAPI et dashboard Streamlit.
PythonReinforcement LearningPPOGymnasiumStable-Baselines3FastAPIStreamlit
Role
Reinforcement learning experimentation, PPO training, API serving, dashboarding, and automated testing.
Outcome
A trained LunarLander-v3 agent with evaluated PPO configurations, saved model artifacts, a FastAPI inference endpoint, and a Streamlit results dashboard.