misza222/ppo-PyramidsTraining
055
1default_settings: null2behaviors:3 Pyramids:4 trainer_type: ppo5 hyperparameters:6 batch_size: 1287 buffer_size: 20488 learning_rate: 0.00039 beta: 0.0110 epsilon: 0.211 lambd: 0.9512 num_epoch: 313 learning_rate_schedule: linear14 beta_schedule: linear15 epsilon_schedule: linear16 network_settings:17 normalize: false18 hidden_units: 51219 num_layers: 220 vis_encode_type: simple21 memory: null22 goal_conditioning_type: hyper23 deterministic: false24 reward_signals:25 extrinsic:26 gamma: 0.9927 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 rnd:37 gamma: 0.9938 strength: 0.0139 network_settings:40 normalize: false41 hidden_units: 6442 num_layers: 343 vis_encode_type: simple44 memory: null45 goal_conditioning_type: hyper46 deterministic: false47 learning_rate: 0.000148 encoding_size: null49 init_path: null50 keep_checkpoints: 551 checkpoint_interval: 50000052 max_steps: 300000053 time_horizon: 12854 summary_freq: 3000055 threaded: false56 self_play: null57 behavioral_cloning: null58env_settings:59 env_path: ./training-envs-executables/linux/Pyramids/Pyramids60 env_args: null61 base_port: 500562 num_envs: 1063 num_areas: 164 seed: -165 max_lifetime_restarts: 1066 restarts_rate_limit_n: 167 restarts_rate_limit_period_s: 6068engine_settings:69 width: 8470 height: 8471 quality_level: 572 time_scale: 2073 target_frame_rate: -174 capture_frame_rate: 6075 no_graphics: true76environment_parameters: null77checkpoint_settings:78 run_id: Pyramids Training79 initialize_from: null80 load_model: false81 resume: false82 force: true83 train_model: false84 inference: false85 results_dir: results86torch_settings:87 device: null88debug: false89