AlphazeroCartpole.json

{
  "name": "singleplayergames",
  "algorithm": "ALPHAZERO",
  "architecture": "Gym",
  
  "args": {
    "num_selfplay_iterations": 100,
    "num_episodes": 20,
    "num_gradient_steps": 40,
    "max_episode_moves": 500,
    "max_trial_moves": 500,
    "pitting": true,
    "pitting_trials": 10,
    "pit_acceptance_ratio": 0.0,
    "dirichlet_alpha": 0.25,
    "exploration_fraction": 0.25,
    "max_buffer_size": 200000,
    "num_MCTS_sims": 11,
    "prioritize": false,
    "prioritize_alpha": 0.5,
    "prioritize_beta": 1,
    "n_steps": 50,
    "c1": 1.25,
    "c2": 19652,
    "gamma": 0.997,

    "minimum_reward": null,
    "maximum_reward": null,

    "checkpoint": "./out/AlphaZeroOut/",
    "load_model": false,
    "load_folder_file": ["./out/AlphaZeroOut/", "latest.pth.tar"],
    "selfplay_buffer_window": 10,

    "temperature_schedule": {
      "method": "stepwise",
      "by_weight_update": true,
      "schedule_points": [[0, 1]]
    }
  },
  
  "net_args": {
    "optimizer": {
      "method": "adam",
      "lr_init": 0.02,
      "momentum": 0.9
    },
    "l2": 1e-4,
    "dropout": 0.1,
    "batch_size": 128,
    "cuda": false,
    "num_channels": 256,
    "num_convs": 2,
    "num_towers": 3,
    "residual_left": 2,
    "residual_right": 0,
    "num_dense": 2,
    "size_dense": 16,
    "activation": "elu",
    "support_size": 20,
    "observation_length": 1
  }
}