Downloads · 30 days
0
q-FrozenLake-v1-8x8-non_slippery
SimingSiming/q-FrozenLake-v1-8x8-non_slippery
q-FrozenLake-v1-8x8-non_slippery is a reinforcement learning model from SimingSiming. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product.
This is a trained model of a Q-Learning agent playing FrozenLake-v1 .
Access
Public
Updated Dec 2, 2022
Repo size
112 KB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.mp4109 KB · 95%
At a glance
- Task
- Reinforcement Learning
- Access
- Public
- Created
- May 27, 2022
- Updated
- Dec 2, 2022
- SHA
- de4cbd01
Model card
From the Hugging Face model README
Q-Learning Agent playing FrozenLake-v1
This is a trained model of a Q-Learning agent playing FrozenLake-v1 .
n_training_episodes = 200000 # Total training episodes <br> learning_rate = 0.8 # Learning rate <br>
Evaluation parameters
n_eval_episodes = 100 # Total number of test episodes <br>
Environment parameters <br>
env_id = "FrozenLake-v1" # Name of the environment <br> max_steps = 100 # Max steps per episode <br> gamma = 0.99 # Discounting rate <br> eval_seed = [] # The evaluation seed of the environment <br>
Exploration parameters <br>
epsilon = 1.0 # Exploration rate <br> max_epsilon = 1.0 # Exploration probability at start <br> min_epsilon = 0.05 # Minimum exploration probability <br> decay_rate = 0.00005 # Exponential decay rate for exploration prob <br>
- Task
- Reinforcement Learning
- Created
- May 27, 2022
- Updated
- Dec 2, 2022