Q-Learning Agent playing FrozenLake-v1
This is a trained model of a **Q-Learning** agent playing **FrozenLake-v1**.
## Model Details
- **Environment**: Taxi-v3
- **Learning Rate**: 0.7
- **Gamma**: 0.95
- **Episodes**: 10000
- **Mean Reward**: 7.60 +/- 2.52
## Usage
```python
import gymnasium as gym
import pickle5 as pickle
# Load the model
with open("q-learning.pkl", "rb") as f:
model = pickle.load(f)
env = gym.make(
"FrozenLake-v1",
is_slippery=model["is_slippery"],
map_name=model["map_name"]
)
# Use the model
state, info = env.reset()
action = np.argmax(model["qtable"][state][:])
```
