Cyber-Machine/ppo-Huggy

1default_settings: null2behaviors:3  Huggy:4    trainer_type: ppo5    hyperparameters:6      batch_size: 20487      buffer_size: 204808      learning_rate: 0.00039      beta: 0.00510      epsilon: 0.211      lambd: 0.9512      num_epoch: 313      learning_rate_schedule: linear14      beta_schedule: linear15      epsilon_schedule: linear16    network_settings:17      normalize: true18      hidden_units: 51219      num_layers: 320      vis_encode_type: simple21      memory: null22      goal_conditioning_type: hyper23      deterministic: false24    reward_signals:25      extrinsic:26        gamma: 0.99527        strength: 1.028        network_settings:29          normalize: false30          hidden_units: 12831          num_layers: 232          vis_encode_type: simple33          memory: null34          goal_conditioning_type: hyper35          deterministic: false36    init_path: null37    keep_checkpoints: 1538    checkpoint_interval: 20000039    max_steps: 200000040    time_horizon: 100041    summary_freq: 5000042    threaded: false43    self_play: null44    behavioral_cloning: null45env_settings:46  env_path: ./trained-envs-executables/linux/Huggy/Huggy47  env_args: null48  base_port: 500549  num_envs: 150  num_areas: 151  seed: -152  max_lifetime_restarts: 1053  restarts_rate_limit_n: 154  restarts_rate_limit_period_s: 6055engine_settings:56  width: 8457  height: 8458  quality_level: 559  time_scale: 2060  target_frame_rate: -161  capture_frame_rate: 6062  no_graphics: true63environment_parameters: null64checkpoint_settings:65  run_id: Huggy66  initialize_from: null67  load_model: false68  resume: false69  force: false70  train_model: false71  inference: false72  results_dir: results73torch_settings:74  device: null75debug: false76