metadata
tags:
- Taxi-v3
- q-learning
- reinforcement-learning
- custom-implementation
model-index:
- name: q-Taxi-v3
results:
- metrics:
- type: mean_reward
value: 7.54 +/- 2.73
name: mean_reward
task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: Taxi-v3
type: Taxi-v3
# 使用**Q-Learning**智能体来玩**Taxi-v3**
这是一个使用**Q-Learning**训练有素的模型玩**Taxi-v3**.
## 用法
```python
model = load_from_hub(repo_id='sun1638650145/q-Taxi-v3', filename='q-learning.pkl')
# 不要忘记检查是否需要添加额外的参数(例如is_slippery=False)
env = gym.make(model['env_id'])
evaluate_agent(env, model['max_steps'], model['n_eval_episodes'], model['qtable'], model['eval_seed'])
```