johnjim0816
add all benchmarks
62e03a2
raw
history blame
5.93 kB
2023-04-02 23:19:50 - r - INFO: - Hyperparameters:
2023-04-02 23:19:50 - r - INFO: - ================================================================================
2023-04-02 23:19:50 - r - INFO: - Name Value Type
2023-04-02 23:19:50 - r - INFO: - env_name LunarLander-v2 <class 'str'>
2023-04-02 23:19:50 - r - INFO: - new_step_api 1 <class 'bool'>
2023-04-02 23:19:50 - r - INFO: - wrapper None <class 'str'>
2023-04-02 23:19:50 - r - INFO: - render 0 <class 'bool'>
2023-04-02 23:19:50 - r - INFO: - algo_name PPO <class 'str'>
2023-04-02 23:19:50 - r - INFO: - mode test <class 'str'>
2023-04-02 23:19:50 - r - INFO: - mp_backend mp <class 'str'>
2023-04-02 23:19:50 - r - INFO: - seed 1 <class 'int'>
2023-04-02 23:19:50 - r - INFO: - device cpu <class 'str'>
2023-04-02 23:19:50 - r - INFO: - train_eps 600 <class 'int'>
2023-04-02 23:19:50 - r - INFO: - test_eps 20 <class 'int'>
2023-04-02 23:19:50 - r - INFO: - eval_eps 10 <class 'int'>
2023-04-02 23:19:50 - r - INFO: - eval_per_episode 5 <class 'int'>
2023-04-02 23:19:50 - r - INFO: - max_steps 1000 <class 'int'>
2023-04-02 23:19:50 - r - INFO: - load_checkpoint 1 <class 'bool'>
2023-04-02 23:19:50 - r - INFO: - load_path Train_LunarLander-v2_PPO_20230402-223154 <class 'str'>
2023-04-02 23:19:50 - r - INFO: - show_fig 0 <class 'bool'>
2023-04-02 23:19:50 - r - INFO: - save_fig 1 <class 'bool'>
2023-04-02 23:19:50 - r - INFO: - ppo_type clip <class 'str'>
2023-04-02 23:19:50 - r - INFO: - continuous 0 <class 'bool'>
2023-04-02 23:19:50 - r - INFO: - gamma 0.99 <class 'float'>
2023-04-02 23:19:50 - r - INFO: - k_epochs 4 <class 'int'>
2023-04-02 23:19:50 - r - INFO: - actor_lr 0.0003 <class 'float'>
2023-04-02 23:19:50 - r - INFO: - critic_lr 0.001 <class 'float'>
2023-04-02 23:19:50 - r - INFO: - eps_clip 0.2 <class 'float'>
2023-04-02 23:19:50 - r - INFO: - entropy_coef 0.01 <class 'float'>
2023-04-02 23:19:50 - r - INFO: - train_batch_size 256 <class 'int'>
2023-04-02 23:19:50 - r - INFO: - sgd_batch_size 32 <class 'int'>
2023-04-02 23:19:50 - r - INFO: - actor_hidden_dim 256 <class 'int'>
2023-04-02 23:19:50 - r - INFO: - critic_hidden_dim 256 <class 'int'>
2023-04-02 23:19:50 - r - INFO: - task_dir /home/zf/Documents/Python/joyrl-offline_off/tasks/Test_LunarLander-v2_PPO_20230402-231950 <class 'str'>
2023-04-02 23:19:50 - r - INFO: - res_dir /home/zf/Documents/Python/joyrl-offline_off/tasks/Test_LunarLander-v2_PPO_20230402-231950/results <class 'str'>
2023-04-02 23:19:50 - r - INFO: - log_dir /home/zf/Documents/Python/joyrl-offline_off/tasks/Test_LunarLander-v2_PPO_20230402-231950/logs <class 'str'>
2023-04-02 23:19:50 - r - INFO: - traj_dir /home/zf/Documents/Python/joyrl-offline_off/tasks/Test_LunarLander-v2_PPO_20230402-231950/traj <class 'str'>
2023-04-02 23:19:50 - r - INFO: - tb_dir /home/zf/Documents/Python/joyrl-offline_off/tasks/Test_LunarLander-v2_PPO_20230402-231950/tb_logs <class 'str'>
2023-04-02 23:19:50 - r - INFO: - ================================================================================
2023-04-02 23:19:50 - r - INFO: - n_states: 8, n_actions: 4
2023-04-02 23:19:50 - r - INFO: - Start testing!
2023-04-02 23:19:50 - r - INFO: - Env: LunarLander-v2, Algorithm: PPO, Device: cpu
2023-04-02 23:19:50 - r - INFO: - Episode: 1/20, Reward: 252.001, Step: 220
2023-04-02 23:19:50 - r - INFO: - Episode: 2/20, Reward: 248.030, Step: 255
2023-04-02 23:19:51 - r - INFO: - Episode: 3/20, Reward: 270.691, Step: 234
2023-04-02 23:19:51 - r - INFO: - Episode: 4/20, Reward: 251.967, Step: 237
2023-04-02 23:19:51 - r - INFO: - Episode: 5/20, Reward: 243.157, Step: 235
2023-04-02 23:19:51 - r - INFO: - Episode: 6/20, Reward: 249.370, Step: 229
2023-04-02 23:19:51 - r - INFO: - Episode: 7/20, Reward: 249.592, Step: 227
2023-04-02 23:19:51 - r - INFO: - Episode: 8/20, Reward: 250.621, Step: 229
2023-04-02 23:19:51 - r - INFO: - Episode: 9/20, Reward: 252.866, Step: 225
2023-04-02 23:19:51 - r - INFO: - Episode: 10/20, Reward: 266.123, Step: 237
2023-04-02 23:19:51 - r - INFO: - Episode: 11/20, Reward: 269.425, Step: 247
2023-04-02 23:19:51 - r - INFO: - Episode: 12/20, Reward: 260.002, Step: 209
2023-04-02 23:19:51 - r - INFO: - Episode: 13/20, Reward: 270.213, Step: 244
2023-04-02 23:19:51 - r - INFO: - Episode: 14/20, Reward: 38.579, Step: 158
2023-04-02 23:19:52 - r - INFO: - Episode: 15/20, Reward: 42.728, Step: 181
2023-04-02 23:19:52 - r - INFO: - Episode: 16/20, Reward: 262.007, Step: 209
2023-04-02 23:19:52 - r - INFO: - Episode: 17/20, Reward: 267.853, Step: 236
2023-04-02 23:19:52 - r - INFO: - Episode: 18/20, Reward: 46.030, Step: 169
2023-04-02 23:19:52 - r - INFO: - Episode: 19/20, Reward: 254.678, Step: 277
2023-04-02 23:19:52 - r - INFO: - Episode: 20/20, Reward: 18.323, Step: 158
2023-04-02 23:19:52 - r - INFO: - Finish testing!