johnjim0816
add all benchmarks
62e03a2
raw
history blame
6.19 kB
2023-04-16 02:26:52 - r - INFO: - Hyperparameters:
2023-04-16 02:26:52 - r - INFO: - ================================================================================
2023-04-16 02:26:52 - r - INFO: - Name Value Type
2023-04-16 02:26:52 - r - INFO: - env_name gym <class 'str'>
2023-04-16 02:26:52 - r - INFO: - new_step_api 1 <class 'bool'>
2023-04-16 02:26:52 - r - INFO: - wrapper None <class 'str'>
2023-04-16 02:26:52 - r - INFO: - render 0 <class 'bool'>
2023-04-16 02:26:52 - r - INFO: - render_mode None <class 'str'>
2023-04-16 02:26:52 - r - INFO: - algo_name SAC <class 'str'>
2023-04-16 02:26:52 - r - INFO: - mode test <class 'str'>
2023-04-16 02:26:52 - r - INFO: - mp_backend mp <class 'str'>
2023-04-16 02:26:52 - r - INFO: - seed 666 <class 'int'>
2023-04-16 02:26:52 - r - INFO: - device cuda <class 'str'>
2023-04-16 02:26:52 - r - INFO: - train_eps 2000 <class 'int'>
2023-04-16 02:26:52 - r - INFO: - test_eps 20 <class 'int'>
2023-04-16 02:26:52 - r - INFO: - eval_eps 10 <class 'int'>
2023-04-16 02:26:52 - r - INFO: - eval_per_episode 5 <class 'int'>
2023-04-16 02:26:52 - r - INFO: - max_steps 800 <class 'int'>
2023-04-16 02:26:52 - r - INFO: - load_checkpoint 1 <class 'bool'>
2023-04-16 02:26:52 - r - INFO: - load_path Train_gym_SAC_20230415-140928 <class 'str'>
2023-04-16 02:26:52 - r - INFO: - show_fig 0 <class 'bool'>
2023-04-16 02:26:52 - r - INFO: - save_fig 1 <class 'bool'>
2023-04-16 02:26:52 - r - INFO: - continous 1 <class 'bool'>
2023-04-16 02:26:52 - r - INFO: - policy_type Gaussian <class 'str'>
2023-04-16 02:26:52 - r - INFO: - lr 0.0004 <class 'float'>
2023-04-16 02:26:52 - r - INFO: - gamma 0.98 <class 'float'>
2023-04-16 02:26:52 - r - INFO: - tau 0.01 <class 'float'>
2023-04-16 02:26:52 - r - INFO: - alpha 0.01 <class 'float'>
2023-04-16 02:26:52 - r - INFO: - automatic_entropy_tuning 0 <class 'bool'>
2023-04-16 02:26:52 - r - INFO: - batch_size 64 <class 'int'>
2023-04-16 02:26:52 - r - INFO: - hidden_dim 256 <class 'int'>
2023-04-16 02:26:52 - r - INFO: - n_epochs 1 <class 'int'>
2023-04-16 02:26:52 - r - INFO: - start_steps 10000 <class 'int'>
2023-04-16 02:26:52 - r - INFO: - target_update_fre 1 <class 'int'>
2023-04-16 02:26:52 - r - INFO: - buffer_size 1000000 <class 'int'>
2023-04-16 02:26:52 - r - INFO: - id BipedalWalker-v3 <class 'str'>
2023-04-16 02:26:52 - r - INFO: - task_dir C:\Users\86153\Desktop\lab\joyrl-offline/tasks/Test_gym_SAC_20230416-022652 <class 'str'>
2023-04-16 02:26:52 - r - INFO: - res_dir C:\Users\86153\Desktop\lab\joyrl-offline/tasks/Test_gym_SAC_20230416-022652/results <class 'str'>
2023-04-16 02:26:52 - r - INFO: - log_dir C:\Users\86153\Desktop\lab\joyrl-offline/tasks/Test_gym_SAC_20230416-022652/logs <class 'str'>
2023-04-16 02:26:52 - r - INFO: - traj_dir C:\Users\86153\Desktop\lab\joyrl-offline/tasks/Test_gym_SAC_20230416-022652/traj <class 'str'>
2023-04-16 02:26:52 - r - INFO: - video_dir C:\Users\86153\Desktop\lab\joyrl-offline/tasks/Test_gym_SAC_20230416-022652/videos <class 'str'>
2023-04-16 02:26:52 - r - INFO: - ================================================================================
2023-04-16 02:26:52 - r - INFO: - action_bound: 1.0
2023-04-16 02:26:52 - r - INFO: - n_states: 24, n_actions: 4
2023-04-16 02:26:54 - r - INFO: - Start testing!
2023-04-16 02:26:54 - r - INFO: - Env: gym, Algorithm: SAC, Device: cuda
2023-04-16 02:26:56 - r - INFO: - Episode: 1/20, Reward: 317.463, Step: 749
2023-04-16 02:26:57 - r - INFO: - Episode: 2/20, Reward: 317.753, Step: 755
2023-04-16 02:26:58 - r - INFO: - Episode: 3/20, Reward: 317.815, Step: 736
2023-04-16 02:26:59 - r - INFO: - Episode: 4/20, Reward: 317.523, Step: 757
2023-04-16 02:27:00 - r - INFO: - Episode: 5/20, Reward: 317.422, Step: 755
2023-04-16 02:27:01 - r - INFO: - Episode: 6/20, Reward: 317.201, Step: 753
2023-04-16 02:27:02 - r - INFO: - Episode: 7/20, Reward: 317.512, Step: 753
2023-04-16 02:27:03 - r - INFO: - Episode: 8/20, Reward: 316.893, Step: 757
2023-04-16 02:27:04 - r - INFO: - Episode: 9/20, Reward: 316.954, Step: 758
2023-04-16 02:27:06 - r - INFO: - Episode: 10/20, Reward: 317.564, Step: 746
2023-04-16 02:27:07 - r - INFO: - Episode: 11/20, Reward: 317.620, Step: 754
2023-04-16 02:27:09 - r - INFO: - Episode: 12/20, Reward: 317.115, Step: 753
2023-04-16 02:27:10 - r - INFO: - Episode: 13/20, Reward: 317.953, Step: 750
2023-04-16 02:27:11 - r - INFO: - Episode: 14/20, Reward: 317.991, Step: 747
2023-04-16 02:27:13 - r - INFO: - Episode: 15/20, Reward: 317.402, Step: 761
2023-04-16 02:27:14 - r - INFO: - Episode: 16/20, Reward: 318.334, Step: 741
2023-04-16 02:27:16 - r - INFO: - Episode: 17/20, Reward: 318.368, Step: 743
2023-04-16 02:27:17 - r - INFO: - Episode: 18/20, Reward: 316.972, Step: 759
2023-04-16 02:27:19 - r - INFO: - Episode: 19/20, Reward: 316.778, Step: 757
2023-04-16 02:27:20 - r - INFO: - Episode: 20/20, Reward: 317.033, Step: 752
2023-04-16 02:27:20 - r - INFO: - Finish testing!