Pong-PLE-v0 reinforce reinforcement-learning custom-implementation deep-rl-class

parameters

pong_hyperparameters = { <br> "h_size": 64,<br> "n_training_episodes": 20000,<br> "n_evaluation_episodes": 10,<br> "max_t": 5000,<br> "gamma": 0.99,<br> "lr": 1e-2,<br> "env_id": env_id,<br> "state_space": s_size,<br> "action_space": a_size,<br> }<br>