Exemple #1
0
        ]
    },
    c.MIN_ACTION: min_action,
    c.MAX_ACTION: max_action,
    c.OBS_DIM: obs_dim,

    # Evaluation
    c.EVALUATION_FREQUENCY: 5000,
    c.EVALUATION_RENDER: False,
    c.EVALUATION_RETURNS: [],
    c.NUM_EVALUATION_EPISODES: 5,

    # Exploration
    c.EXPLORATION_STEPS: 1000,
    c.EXPLORATION_STRATEGY: UniformContinuousAgent(min_action,
                                                   max_action,
                                                   np.random.RandomState(seed)),
    
    # General
    c.DEVICE: device,
    c.SEED: seed,

    # Load
    c.LOAD_MODEL: False,

    # Logging
    c.PRINT_INTERVAL: 5000,
    c.SAVE_INTERVAL: 1000000,

    # Model
    c.MODEL_SETTING: {
    np.ones(action_dim),

    # Evaluation
    c.EVALUATION_FREQUENCY:
    5000,
    c.EVALUATION_RENDER:
    False,
    c.EVALUATION_RETURNS: [],
    c.NUM_EVALUATION_EPISODES:
    5,

    # Exploration
    c.EXPLORATION_STEPS:
    10000,
    c.EXPLORATION_STRATEGY:
    UniformContinuousAgent(min_action, max_action,
                           np.random.RandomState(seed)),

    # General
    c.DEVICE:
    device,
    c.SEED:
    seed,

    # Load
    c.LOAD_MODEL:
    False,

    # Logging
    c.PRINT_INTERVAL:
    5000,
    c.SAVE_INTERVAL: