berkde/sarsa-lambda-WindyGridworld-v0

Q-Learning Agent playing WindyCliffEnv-v0-8x8-noSlippery

This is a trained model of a Q-Learning agent playing WindyCliffEnv-v0-8x8-noSlippery.

Usage

from huggingface_hub import hf_hub_download
import pickle, gymnasium as gym, numpy as np

# Download Q-table
pkl_path = hf_hub_download(repo_id="berkde/sarsa-lambda-WindyGridworld-v0", filename="q-learning.pkl")
with open(pkl_path, "rb") as f:
    model = pickle.load(f)

env = gym.make(
    model["env_id"],
    map_name=model.get("map_name"),
    is_slippery=model.get("slippery", True),
    render_mode="human",
)

state, _ = env.reset()
for _ in range(model["max_steps"]):
    action = np.argmax(model["qtable"][state])
    state, reward, terminated, truncated, _ = env.step(action)
    if terminated or truncated:
        break