Reinforce-PongPolicyGradient / hyperparameters.json
jackoyoungblood's picture
Reinforce-PongPolGrad-2200 training episodes
dbfa02e
raw
history blame
175 Bytes
{"h_size": 64, "n_training_episodes": 2200, "n_evaluation_episodes": 13, "max_t": 5000, "gamma": 0.96, "lr": 0.1, "env_id": "CartPole-v1", "state_space": 4, "action_space": 2}