MattStammers's picture
Update README.md
ff9c18a
metadata
tags:
  - CartPole-v1
  - ppo
  - deep-reinforcement-learning
  - reinforcement-learning
  - custom-implementation
  - deep-rl-course
model-index:
  - name: PPO
    results:
      - task:
          type: reinforcement-learning
          name: reinforcement-learning
        dataset:
          name: CartPole-v1
          type: CartPole-v1
        metrics:
          - type: mean_reward
            value: 414.70 +/- 136.64
            name: mean_reward
            verified: false

PPO Agent Playing CartPole-v1

This is a trained model of a PPO agent playing CartPole-v1.

Hyperparameters

This was trained with a decaying learning rate of 25e-5, clip-coef of 0.1 and 1 million timesteps.

Still not perfect but that was not the point. You can follow the full journey here for more info: https://github.com/MattStammers/PPO_Lander_Implementation