PPO LunarLander-v2
Trained from scratch in the Hugging Face Unit 1 notebook with AI assistance. Training seed: 42. Steps: 1015808. Evaluation: 20 deterministic episodes, evaluation environment seed 2026. Course score (mean minus standard deviation): 215.72.
Load
Use the dependencies in requirements.txt, then:
from stable_baselines3 import PPO
model = PPO.load("ppo-LunarLander-v2.zip", device="cpu")
- Downloads last month
- 16
Evaluation results
- mean_reward on LunarLander-v2self-reported259.72 +/- 44.00