FrozenLake-v1-4x4 q-learning reinforcement-learning custom-implementation

Tricky

Slippery devil but he does an ok job.

Hint: Overtraining will not help you here

Usage

model = load_from_hub(repo_id="MattStammers/q-FrozenLake-v1-4x4-Slippery", filename="q-learning.pkl")

Then I recommend you use the code provided in the hands-on notebook to run the model