HumanCompatibleAI/sac-seals-Humanoid-v1

SAC Agent playing seals/Humanoid-v1

This is a trained model of a SAC agent playing seals/Humanoid-v1 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 sac --env seals/Humanoid-v1 -orga HumanCompatibleAI -f logs/
python -m rl_zoo3.enjoy --algo sac --env seals/Humanoid-v1  -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 sac --env seals/Humanoid-v1 -orga HumanCompatibleAI -f logs/
python -m rl_zoo3.enjoy --algo sac --env seals/Humanoid-v1  -f logs/

Training (with the RL Zoo)

python -m rl_zoo3.train --algo sac --env seals/Humanoid-v1 -f logs/
# Upload the model and generate video (when possible)
python -m rl_zoo3.push_to_hub --algo sac --env seals/Humanoid-v1 -f logs/ -orga HumanCompatibleAI

Hyperparameters

OrderedDict([('batch_size', 64),
             ('buffer_size', 100000),
             ('gamma', 0.98),
             ('learning_rate', 4.426351861707874e-05),
             ('learning_starts', 20000),
             ('n_timesteps', 2000000.0),
             ('policy', 'MlpPolicy'),
             ('policy_kwargs',
              {'log_std_init': -0.1034412732183072,
               'net_arch': [400, 300],
               'use_sde': False}),
             ('tau', 0.08),
             ('train_freq', 8),
             ('normalize', False)])

Environment Arguments

{'render_mode': 'rgb_array'}