behaviors:
  VisualPyramids:
    trainer: sac
    batch_size: 64
    buffer_size: 500000
    buffer_init_steps: 1000
    hidden_units: 256
    init_entcoef: 0.01
    learning_rate: 0.0003
    learning_rate_schedule: constant
    max_steps: 1.0e7
    memory_size: 128
    normalize: false
    steps_per_update: 10
    num_layers: 1
    time_horizon: 128
    sequence_length: 64
    summary_freq: 10000
    tau: 0.01
    use_recurrent: false
    vis_encode_type: simple
    reward_signals:
      extrinsic:
        strength: 2.0
        gamma: 0.99
      gail:
        strength: 0.02
        gamma: 0.99
        encoding_size: 128
        use_actions: true
        demo_path: Project/Assets/ML-Agents/Examples/Pyramids/Demos/ExpertPyramid.demo