2022-12-17 20:42:14 - r - INFO: - Hyperparameters: 2022-12-17 20:42:14 - r - INFO: - ================================================================================ 2022-12-17 20:42:14 - r - INFO: - Name Value Type 2022-12-17 20:42:14 - r - INFO: - env_name CartPole-v1 2022-12-17 20:42:14 - r - INFO: - new_step_api 1 2022-12-17 20:42:14 - r - INFO: - wrapper None 2022-12-17 20:42:14 - r - INFO: - render 0 2022-12-17 20:42:14 - r - INFO: - algo_name PPO 2022-12-17 20:42:14 - r - INFO: - mode test 2022-12-17 20:42:14 - r - INFO: - seed 1 2022-12-17 20:42:14 - r - INFO: - device cuda 2022-12-17 20:42:14 - r - INFO: - train_eps 200 2022-12-17 20:42:14 - r - INFO: - test_eps 10 2022-12-17 20:42:14 - r - INFO: - eval_eps 10 2022-12-17 20:42:14 - r - INFO: - eval_per_episode 5 2022-12-17 20:42:14 - r - INFO: - max_steps 200 2022-12-17 20:42:14 - r - INFO: - load_checkpoint 1 2022-12-17 20:42:14 - r - INFO: - load_path Train_CartPole-v1_PPO_20221217-204003 2022-12-17 20:42:14 - r - INFO: - show_fig 0 2022-12-17 20:42:14 - r - INFO: - save_fig 1 2022-12-17 20:42:14 - r - INFO: - ppo_type kl 2022-12-17 20:42:14 - r - INFO: - continuous 0 2022-12-17 20:42:14 - r - INFO: - gamma 0.99 2022-12-17 20:42:14 - r - INFO: - k_epochs 4 2022-12-17 20:42:14 - r - INFO: - actor_lr 0.0003 2022-12-17 20:42:14 - r - INFO: - critic_lr 0.001 2022-12-17 20:42:14 - r - INFO: - eps_clip 0.2 2022-12-17 20:42:14 - r - INFO: - entropy_coef 0.01 2022-12-17 20:42:14 - r - INFO: - train_batch_size 100 2022-12-17 20:42:14 - r - INFO: - sgd_batch_size 64 2022-12-17 20:42:14 - r - INFO: - actor_hidden_dim 256 2022-12-17 20:42:14 - r - INFO: - critic_hidden_dim 256 2022-12-17 20:42:14 - r - INFO: - kl_alpha 2 2022-12-17 20:42:14 - r - INFO: - kl_beta 1.5 2022-12-17 20:42:14 - r - INFO: - kl_lambda 0.5 2022-12-17 20:42:14 - r - INFO: - kl_target 0.01 2022-12-17 20:42:14 - r - INFO: - ================================================================================ 2022-12-17 20:42:15 - r - INFO: - n_states: 4, n_actions: 2 2022-12-17 20:42:16 - r - INFO: - Start testing! 2022-12-17 20:42:16 - r - INFO: - Env: CartPole-v1, Algorithm: PPO, Device: cuda 2022-12-17 20:42:17 - r - INFO: - Episode: 1/10, Reward: 200.000, Step: 200 2022-12-17 20:42:17 - r - INFO: - Episode: 2/10, Reward: 200.000, Step: 200 2022-12-17 20:42:18 - r - INFO: - Episode: 3/10, Reward: 200.000, Step: 200 2022-12-17 20:42:18 - r - INFO: - Episode: 4/10, Reward: 200.000, Step: 200 2022-12-17 20:42:18 - r - INFO: - Episode: 5/10, Reward: 200.000, Step: 200 2022-12-17 20:42:19 - r - INFO: - Episode: 6/10, Reward: 200.000, Step: 200 2022-12-17 20:42:19 - r - INFO: - Episode: 7/10, Reward: 200.000, Step: 200 2022-12-17 20:42:19 - r - INFO: - Episode: 8/10, Reward: 200.000, Step: 200 2022-12-17 20:42:19 - r - INFO: - Episode: 9/10, Reward: 200.000, Step: 200 2022-12-17 20:42:20 - r - INFO: - Episode: 10/10, Reward: 200.000, Step: 200 2022-12-17 20:42:20 - r - INFO: - Finish testing!