experiment_421_config.json

{
    "human_demonstration_buffer_builder": {
        "keyboard_map_filename": null,
        "set_default_noop": null,
        "path_to_store_game": "config/421_hbuffer.pickle"
    },
    "game": {
        "env": "CartPole-v1",
        "render": null
    },
    "random_seed": {
        "np_random_seed": 0,
        "torch_manual_seed": 0,
        "env_seed": 0
    },
    "muzero": {
        "model_structure": "mlp_model",
        "state_space_dimensions": 31,
        "hidden_layer_dimensions": 64,
        "number_of_hidden_layer": 0,
        "k_hypothetical_steps": 10,
        "optimizer": "adam",
        "lr_scheduler": null,
        "learning_rate": 0.01,
        "loss_type": "general",
        "num_of_epoch": 10000,
        "device": "cpu",
        "load": false,
        "use_amp": false,
        "scaler_on": false,
        "bin_method": "uniform_bin",
        "bin_decomposition_number": 10,
        "priority_scale": 0.5,
        "rescale_value_loss": 1.0
    },
    "replaybuffer": {
        "window_size": 500,
        "batch_size": 128,
        "td_steps": 50,
        "game_sampling": "priority",
        "position_sampling": "priority",
        "reanalyse_fraction": 0.0,
        "reanalyse_fraction_mode": "chance"
    },
    "monte_carlo_tree_search": {
        "pb_c_base": 19652,
        "pb_c_init": 1.25,
        "discount": 0.997,
        "root_dirichlet_alpha": 0.25,
        "root_exploration_fraction": 0.25,
        "num_simulations": 11,
        "maxium_action_sample": 2,
        "number_of_player": 1,
        "custom_loop": null
    },
    "gameplay": {
        "limit_of_game_play": 500
    },
    "learning_cycle": {
        "number_of_iteration": 10000,
        "number_of_self_play_before_training": 1,
        "number_of_training_before_self_play": 1,
        "temperature_type": "static_temperature",
        "model_tag_number": 421,
        "verbose": true,
        "number_of_worker_selfplay": 0
    },
    "play_game_from_checkpoint": {
        "model_tag": 421,
        "model_device": "cpu",
        "mcts_with_or_without_dirichlet_noise": true,
        "number_of_monte_carlo_tree_search_simulation": 11,
        "temperature": 0,
        "game_iter": 500,
        "slow_mo_in_second": 0.0,
        "render": null,
        "verbose": true
    }
}