{
    "env_name": "Valkyrie-5",
    "policy_name": "DqnPolicy",
    "policy_params": {
        "batch_size": 128,
        "layer_sizes": [
            64,
            64
        ],
        "double_q": true,
        "dueling": false,
        "model_type": "dense",
        "step_size": 1
    },
    "train_params": {
        "lr": 0.001,
        "epsilon": 1.0,
        "epsilon_final": 0.02,
        "warmup_episodes": 500,
        "log_every_episode": 10,
        "n_episodes": 2000,
        "target_update_every_step": 10,
        "update_per_sample": 1,
        "episode_length": 30
    }
}