README.md

metadata

library_name: stable-baselines3
tags:
  - MiniGrid-KeyCorridorS3R1-v0
  - deep-reinforcement-learning
  - reinforcement-learning
  - stable-baselines3
model-index:
  - name: PPO
    results:
      - task:
          type: reinforcement-learning
          name: reinforcement-learning
        dataset:
          name: MiniGrid-KeyCorridorS3R1-v0
          type: MiniGrid-KeyCorridorS3R1-v0
        metrics:
          - type: mean_reward
            value: 0.95 +/- 0.00
            name: mean_reward
            verified: false

PPO Agent playing MiniGrid-KeyCorridorS3R1-v0

This is a trained model of a PPO agent playing MiniGrid-KeyCorridorS3R1-v0 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 ppo --env MiniGrid-KeyCorridorS3R1-v0 -orga sb3 -f logs/
python -m rl_zoo3.enjoy --algo ppo --env MiniGrid-KeyCorridorS3R1-v0  -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 MiniGrid-KeyCorridorS3R1-v0 -orga sb3 -f logs/
python -m rl_zoo3.enjoy --algo ppo --env MiniGrid-KeyCorridorS3R1-v0  -f logs/

Training (with the RL Zoo)

python -m rl_zoo3.train --algo ppo --env MiniGrid-KeyCorridorS3R1-v0 -f logs/
# Upload the model and generate video (when possible)
python -m rl_zoo3.push_to_hub --algo ppo --env MiniGrid-KeyCorridorS3R1-v0 -f logs/ -orga sb3

Hyperparameters

OrderedDict([('batch_size', 64),
             ('clip_range', 0.2),
             ('ent_coef', 0.0),
             ('env_wrapper', 'gym_minigrid.wrappers.FlatObsWrapper'),
             ('gae_lambda', 0.95),
             ('gamma', 0.99),
             ('learning_rate', 0.00025),
             ('n_envs', 8),
             ('n_epochs', 10),
             ('n_steps', 128),
             ('n_timesteps', 500000.0),
             ('normalize', True),
             ('policy', 'MlpPolicy'),
             ('normalize_kwargs', {'norm_obs': True, 'norm_reward': False})])