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

<br> pixelcopter_hyperparameters = { <br> "h_size": 64, <br> "n_training_episodes": 50000, <br> "n_evaluation_episodes": 10, <br> "max_t": 10000, <br> "gamma": 0.99, <br> "lr": 1e-4, <br> "env_id": env_id, <br> "state_space": s_size, <br> "action_space": a_size, <br> }<br>