lsdyna/ppo-SnowballTarget

1default_settings: null2behaviors:3  SnowballTarget:4    trainer_type: ppo5    hyperparameters:6      batch_size: 1287      buffer_size: 20488      learning_rate: 0.00039      beta: 0.00510      epsilon: 0.211      lambd: 0.9512      num_epoch: 313      shared_critic: false14      learning_rate_schedule: linear15      beta_schedule: linear16      epsilon_schedule: linear17    checkpoint_interval: 5000018    network_settings:19      normalize: false20      hidden_units: 25621      num_layers: 222      vis_encode_type: simple23      memory: null24      goal_conditioning_type: hyper25      deterministic: false26    reward_signals:27      extrinsic:28        gamma: 0.9929        strength: 1.030        network_settings:31          normalize: false32          hidden_units: 12833          num_layers: 234          vis_encode_type: simple35          memory: null36          goal_conditioning_type: hyper37          deterministic: false38    init_path: null39    keep_checkpoints: 1040    even_checkpoints: false41    max_steps: 20000042    time_horizon: 6443    summary_freq: 1000044    threaded: false45    self_play: null46    behavioral_cloning: null47env_settings:48  env_path: ./training-envs-executables/linux/SnowballTarget/SnowballTarget49  env_args: null50  base_port: 500551  num_envs: 152  num_areas: 153  timeout_wait: 6054  seed: -155  max_lifetime_restarts: 1056  restarts_rate_limit_n: 157  restarts_rate_limit_period_s: 6058engine_settings:59  width: 8460  height: 8461  quality_level: 562  time_scale: 2063  target_frame_rate: -164  capture_frame_rate: 6065  no_graphics: true66  no_graphics_monitor: false67environment_parameters: null68checkpoint_settings:69  run_id: SnowballTarget170  initialize_from: null71  load_model: false72  resume: false73  force: false74  train_model: false75  inference: false76  results_dir: results77torch_settings:78  device: null79debug: false80