Edit model card

ARS Agent playing BipedalWalkerHardcore-v3

This is a trained model of a ARS agent playing BipedalWalkerHardcore-v3 using the stable-baselines3 library and the RL Zoo.

The RL Zoo is a training framework for Stable Baselines3 reinforcement learning agents, with hyperparameter optimization and pre-trained agents included.

Usage (with SB3 RL Zoo)

RL Zoo: https://github.com/DLR-RM/rl-baselines3-zoo
SB3: https://github.com/DLR-RM/stable-baselines3
SB3 Contrib: https://github.com/Stable-Baselines-Team/stable-baselines3-contrib

Install the RL Zoo (with SB3 and SB3-Contrib):

pip install rl_zoo3
# Download model and save it into the logs/ folder
python -m rl_zoo3.load_from_hub --algo ars --env BipedalWalkerHardcore-v3 -orga qgallouedec -f logs/
python -m rl_zoo3.enjoy --algo ars --env BipedalWalkerHardcore-v3  -f logs/

If you installed the RL Zoo3 via pip (pip install rl_zoo3), from anywhere you can do:

python -m rl_zoo3.load_from_hub --algo ars --env BipedalWalkerHardcore-v3 -orga qgallouedec -f logs/
python -m rl_zoo3.enjoy --algo ars --env BipedalWalkerHardcore-v3  -f logs/

Training (with the RL Zoo)

python -m rl_zoo3.train --algo ars --env BipedalWalkerHardcore-v3 -f logs/
# Upload the model and generate video (when possible)
python -m rl_zoo3.push_to_hub --algo ars --env BipedalWalkerHardcore-v3 -f logs/ -orga qgallouedec

Hyperparameters

OrderedDict([('alive_bonus_offset', -0.1),
             ('delta_std', 0.0075),
             ('learning_rate', 0.02),
             ('n_delta', 64),
             ('n_envs', 1),
             ('n_timesteps', 500000000.0),
             ('n_top', 32),
             ('normalize', 'dict(norm_obs=True, norm_reward=False)'),
             ('policy', 'MlpPolicy'),
             ('policy_kwargs', 'dict(net_arch=[16])'),
             ('normalize_kwargs', {'norm_obs': True, 'norm_reward': False})])
Downloads last month
5
Video Preview
loading

Evaluation results

  • mean_reward on BipedalWalkerHardcore-v3
    self-reported
    -85.86 +/- 5.99