params.json

{
    "env_config": {
        "env_id": "Pendulum-v1",
        "env_kwargs": {},
        "max_steps": null,
        "normalize_obs": false,
        "use_image": false,
        "vector_env_num": 6,
        "use_multi_processing": true,
        "image_shape": null,
        "frame_stack": 1,
        "frame_skip": 1,
        "training_render_mode": null
    },
    "device": "cpu",
    "learning_rate": 0.0003,
    "gamma": 0.99,
    "checkpoint_pathname": "",
    "max_grad_norm": 0.5,
    "log_interval": 100,
    "eval_episodes": 50,
    "eval_random_seed": 42,
    "eval_video_num": 10,
    "total_steps": 240000,
    "hidden_sizes": [
        256,
        256
    ],
    "critic_lr": 0.0003,
    "replay_buffer_capacity": 240000,
    "batch_size": 128,
    "update_start_step": 20000,
    "policy_delay": 2,
    "policy_noise": 0.2,
    "noise_clip": 0.5,
    "exploration_noise": {
        "_type": "LinearSchedule",
        "_module": "practice.utils_for_coding.scheduler_utils",
        "_start_e": 0.3,
        "_end_e": 0.0,
        "_duration": 10000,
        "_start_t": 0
    },
    "max_action": 2.0,
    "tau": 0.005
}