a2c_panda_reach / results.json
loqmen's picture
Upload du modèle A2C entraîné sur PandaReachJointsDense-v3 avec 500000 timesteps
54fff04 verified
{"mean_reward": -0.2924077556468546, "std_reward": 0.13601033423696432, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2025-03-20T19:53:47.690614"}