sb3 / a2c-BipedalWalkerHardcore-v3

A2C Agent playing BipedalWalkerHardcore-v3

This is a trained model of a A2C 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

# Download model and save it into the logs/ folder
python -m rl_zoo3.load_from_hub --algo a2c --env BipedalWalkerHardcore-v3 -orga sb3 -f logs/
python enjoy.py --algo a2c --env BipedalWalkerHardcore-v3  -f logs/

Training (with the RL Zoo)

python train.py --algo a2c --env BipedalWalkerHardcore-v3 -f logs/
# Upload the model and generate video (when possible)
python -m rl_zoo3.push_to_hub --algo a2c --env BipedalWalkerHardcore-v3 -f logs/ -orga sb3

Hyperparameters

OrderedDict([('ent_coef', 0.001),
             ('gae_lambda', 0.9),
             ('gamma', 0.99),
             ('learning_rate', 'lin_0.0008'),
             ('max_grad_norm', 0.5),
             ('n_envs', 32),
             ('n_steps', 8),
             ('n_timesteps', 200000000.0),
             ('normalize', True),
             ('normalize_advantage', False),
             ('policy', 'MlpPolicy'),
             ('policy_kwargs', 'dict(log_std_init=-2, ortho_init=False)'),
             ('use_rms_prop', True),
             ('use_sde', True),
             ('vf_coef', 0.4),
             ('normalize_kwargs', {'norm_obs': True, 'norm_reward': False})])