ByteExplorer/ppo-Pyramids
0133
1default_settings: null2behaviors:3 Pyramids:4 trainer_type: ppo5 hyperparameters:6 batch_size: 2567 buffer_size: 40968 learning_rate: 0.0039 beta: 0.0110 epsilon: 0.211 lambd: 0.9512 num_epoch: 313 shared_critic: false14 learning_rate_schedule: linear15 beta_schedule: linear16 epsilon_schedule: linear17 checkpoint_interval: 50000018 network_settings:19 normalize: false20 hidden_units: 51221 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 rnd:39 gamma: 0.9940 strength: 0.0141 network_settings:42 normalize: false43 hidden_units: 6444 num_layers: 345 vis_encode_type: simple46 memory: null47 goal_conditioning_type: hyper48 deterministic: false49 learning_rate: 0.000150 encoding_size: null51 init_path: null52 keep_checkpoints: 553 even_checkpoints: false54 max_steps: 5000000055 time_horizon: 12856 summary_freq: 3000057 threaded: false58 self_play: null59 behavioral_cloning: null60env_settings:61 env_path: ./training-envs-executables/linux/Pyramids/Pyramids62 env_args: null63 base_port: 500564 num_envs: 165 num_areas: 166 seed: -167 max_lifetime_restarts: 1068 restarts_rate_limit_n: 169 restarts_rate_limit_period_s: 6070engine_settings:71 width: 8472 height: 8473 quality_level: 574 time_scale: 2075 target_frame_rate: -176 capture_frame_rate: 6077 no_graphics: true78environment_parameters: null79checkpoint_settings:80 run_id: Pyramids Training81 initialize_from: null82 load_model: false83 resume: false84 force: true85 train_model: false86 inference: false87 results_dir: results88torch_settings:89 device: null90debug: false91