README.md

metadata

library_name: stable-baselines3
tags:
  - MountainCar-v0
  - deep-reinforcement-learning
  - reinforcement-learning
  - stable-baselines3
model-index:
  - name: PPO
    results:
      - task:
          type: reinforcement-learning
          name: reinforcement-learning
        dataset:
          name: MountainCar-v0
          type: MountainCar-v0
        metrics:
          - type: mean_reward
            value: '-116.20 +/- 1.83'
            name: mean_reward
            verified: false

PPO Agent playing MountainCar-v0

This is a trained model of a PPO agent playing MountainCar-v0 using the stable-baselines3 library.

Model Details

- Model Name: ppo-MountainCar-v0
- Model Type: Proximal Policy Optimization (PPO)
- Policy Architecture: MultiLayerPerceptron (MLPPolicy)
- Environment: MountainCar-v0

• Training Data: The model was trained using three consecutive training sessions:
  • First training session: Total timesteps = 1,000,000
  • Second training session: Total timesteps = 500,000
  • Third training session: Total timesteps = 500,000

Model Parameters

- n_steps: 2048
- batch_size: 64
- n_epochs: 8
- gamma: 0.999
- gae_lambda: 0.95
- ent_coef: 0.01
- max_grad_norm: 0.5
- Verbose: Enabled (Verbose level = 1)