liajun/ppo-PandaReach-v3

PPO Agent playing PandaReach-v3

This is a trained model of a PPO agent playing PandaReach-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 SBX (SB3 + Jax): https://github.com/araffin/sbx

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 ppo --env PandaReach-v3 -orga liajun -f logs/
python -m rl_zoo3.enjoy --algo ppo --env PandaReach-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 ppo --env PandaReach-v3 -orga liajun -f logs/
python -m rl_zoo3.enjoy --algo ppo --env PandaReach-v3  -f logs/

Training (with the RL Zoo)

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

Hyperparameters

OrderedDict([('batch_size', 512),
             ('clip_range', 0.2),
             ('ent_coef', 0.00066),
             ('gae_lambda', 0.9978),
             ('gamma', 0.99779),
             ('learning_rate', 0.00044),
             ('max_grad_norm', 0.77),
             ('n_envs', 16),
             ('n_epochs', 10),
             ('n_steps', 2048),
             ('n_timesteps', 1000000.0),
             ('normalize', True),
             ('policy', 'MultiInputPolicy'),
             ('policy_kwargs',
              'dict( net_arch=[256, 256], activation_fn=nn.ReLU )'),
             ('normalize_kwargs', {'norm_obs': True, 'norm_reward': False})])

Environment Arguments

{'render_mode': 'rgb_array'}