behaviors:
  CrawlerStatic:
    trainer: ppo
    batch_size: 2024
    beta: 0.005
    buffer_size: 20240
    epsilon: 0.2
    hidden_units: 512
    lambd: 0.95
    learning_rate: 0.0003
    max_steps: 1e7
    memory_size: 256
    normalize: true
    num_epoch: 3
    num_layers: 3
    time_horizon: 1000
    sequence_length: 64
    summary_freq: 30000
    use_recurrent: false
    reward_signals:
      gail:
        strength: 1.0
        gamma: 0.99
        encoding_size: 128
        demo_path: Project/Assets/ML-Agents/Examples/Crawler/Demos/ExpertCrawlerSta.demo
    behavioral_cloning:
      demo_path: Project/Assets/ML-Agents/Examples/Crawler/Demos/ExpertCrawlerSta.demo
      strength: 0.5
      steps: 50000