Q-Learning Agent playing FrozenLake-v1

This is a trained model of a Q-Learning agent playing FrozenLake-v1 . Compared to the 4x4 training model, I increased max_steps to increase likelyhood of finding the goal during exploration. That still did not give a good solution. Then I decreased decay rate to allow for more exploration and increased learning rate. But mostly it was the lower decay rate that improved the result.

Usage

model = load_from_hub(repo_id="harshil128/q-FrozenLake-v1-8x8-noSlippery", filename="q-learning.pkl")

# Don't forget to check if you need to add additional attributes (is_slippery=False etc)
env = gym.make(model["env_id"])

evaluate_agent(env, model["max_steps"], model["n_eval_episodes"], model["qtable"], model["eval_seed"])