sgoodfriend's picture
Longer time horizon
88cd07d
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.327528178691864,
"min": 0.327528178691864,
"max": 1.4502936601638794,
"count": 50
},
"Pyramids.Policy.Entropy.sum": {
"value": 9951.6162109375,
"min": 9951.6162109375,
"max": 46409.3984375,
"count": 50
},
"Pyramids.Step.mean": {
"value": 1499905.0,
"min": 29912.0,
"max": 1499905.0,
"count": 50
},
"Pyramids.Step.sum": {
"value": 1499905.0,
"min": 29912.0,
"max": 1499905.0,
"count": 50
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.7862254977226257,
"min": -0.0986347645521164,
"max": 0.7862254977226257,
"count": 50
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 140.73435974121094,
"min": -11.934806823730469,
"max": 140.73435974121094,
"count": 50
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": -0.021666787564754486,
"min": -0.028470776975154877,
"max": 0.2771379351615906,
"count": 50
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": -3.878355026245117,
"min": -4.783090591430664,
"max": 33.256553649902344,
"count": 50
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06795475266324207,
"min": 0.06488773596184995,
"max": 0.0760969590461734,
"count": 50
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9513665372853889,
"min": 0.4749415263652549,
"max": 1.0305533156822684,
"count": 50
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.01551100939516492,
"min": 0.0009322263464162465,
"max": 0.016503496621601946,
"count": 50
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.2171541315323089,
"min": 0.010366753799227942,
"max": 0.22684459496111442,
"count": 50
},
"Pyramids.Policy.LearningRate.mean": {
"value": 3.172570371080951e-06,
"min": 3.172570371080951e-06,
"max": 0.00029676708679192377,
"count": 50
},
"Pyramids.Policy.LearningRate.sum": {
"value": 4.441598519513331e-05,
"min": 4.441598519513331e-05,
"max": 0.0033170318943227996,
"count": 50
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10105749047619049,
"min": 0.10105749047619049,
"max": 0.19892236190476195,
"count": 50
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4148048666666668,
"min": 1.3368084666666666,
"max": 2.5056772,
"count": 50
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00011564329857142854,
"min": 0.00011564329857142854,
"max": 0.009892343954285714,
"count": 50
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0016190061799999996,
"min": 0.0016190061799999996,
"max": 0.11059715228,
"count": 50
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.007556379772722721,
"min": 0.007414631079882383,
"max": 0.46072664856910706,
"count": 50
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.10578931868076324,
"min": 0.09769266843795776,
"max": 3.2250864505767822,
"count": 50
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 231.448,
"min": 231.448,
"max": 999.0,
"count": 50
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 28931.0,
"min": 15850.0,
"max": 33465.0,
"count": 50
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.7674719789028168,
"min": -1.0000000596046448,
"max": 1.7674719789028168,
"count": 50
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 220.9339973628521,
"min": -28.582001954317093,
"max": 225.87839779257774,
"count": 50
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.7674719789028168,
"min": -1.0000000596046448,
"max": 1.7674719789028168,
"count": 50
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 220.9339973628521,
"min": -28.582001954317093,
"max": 225.87839779257774,
"count": 50
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.018012536310357975,
"min": 0.018012536310357975,
"max": 7.0311954423137335,
"count": 50
},
"Pyramids.Policy.RndReward.sum": {
"value": 2.2515670387947466,
"min": 2.2515670387947466,
"max": 161.71749517321587,
"count": 50
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 50
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 50
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1673509934",
"python_version": "3.8.16 (default, Dec 7 2022, 01:12:13) \n[GCC 7.5.0]",
"command_line_arguments": "/usr/local/bin/mlagents-learn ./config/ppo/PyramidsRND.yaml --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids2 --no-graphics",
"mlagents_version": "0.29.0.dev0",
"mlagents_envs_version": "0.29.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "1.8.1+cu102",
"numpy_version": "1.21.6",
"end_time_seconds": "1673512371"
},
"total": 2436.2943086840005,
"count": 1,
"self": 0.3218447830004152,
"children": {
"run_training.setup": {
"total": 0.1023755930000334,
"count": 1,
"self": 0.1023755930000334
},
"TrainerController.start_learning": {
"total": 2435.870088308,
"count": 1,
"self": 1.630570045843342,
"children": {
"TrainerController._reset_env": {
"total": 6.004141104000155,
"count": 1,
"self": 6.004141104000155
},
"TrainerController.advance": {
"total": 2428.1481303021565,
"count": 96814,
"self": 1.7727840980787732,
"children": {
"env_step": {
"total": 1485.060660868984,
"count": 96814,
"self": 1336.6898835476986,
"children": {
"SubprocessEnvManager._take_step": {
"total": 147.30312773816877,
"count": 96814,
"self": 6.095399021332014,
"children": {
"TorchPolicy.evaluate": {
"total": 141.20772871683675,
"count": 93858,
"self": 47.62053048175494,
"children": {
"TorchPolicy.sample_actions": {
"total": 93.58719823508181,
"count": 93858,
"self": 93.58719823508181
}
}
}
}
},
"workers": {
"total": 1.0676495831166903,
"count": 96814,
"self": 0.0,
"children": {
"worker_root": {
"total": 2432.3293087098527,
"count": 96814,
"is_parallel": true,
"self": 1224.0753513419404,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.001793319999706,
"count": 1,
"is_parallel": true,
"self": 0.0006566319998455583,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011366879998604418,
"count": 8,
"is_parallel": true,
"self": 0.0011366879998604418
}
}
},
"UnityEnvironment.step": {
"total": 0.03690019999976357,
"count": 1,
"is_parallel": true,
"self": 0.0005673259984178003,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0005458120003822842,
"count": 1,
"is_parallel": true,
"self": 0.0005458120003822842
},
"communicator.exchange": {
"total": 0.033989833000305225,
"count": 1,
"is_parallel": true,
"self": 0.033989833000305225
},
"steps_from_proto": {
"total": 0.0017972290006582625,
"count": 1,
"is_parallel": true,
"self": 0.0004644370001187781,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0013327920005394844,
"count": 8,
"is_parallel": true,
"self": 0.0013327920005394844
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1208.2539573679123,
"count": 96813,
"is_parallel": true,
"self": 24.264287808257905,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 18.58284851092958,
"count": 96813,
"is_parallel": true,
"self": 18.58284851092958
},
"communicator.exchange": {
"total": 1072.1948206297984,
"count": 96813,
"is_parallel": true,
"self": 1072.1948206297984
},
"steps_from_proto": {
"total": 93.21200041892644,
"count": 96813,
"is_parallel": true,
"self": 21.30576312638459,
"children": {
"_process_rank_one_or_two_observation": {
"total": 71.90623729254185,
"count": 774504,
"is_parallel": true,
"self": 71.90623729254185
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 941.3146853350936,
"count": 96814,
"self": 3.4799849491882924,
"children": {
"process_trajectory": {
"total": 194.82750869390202,
"count": 96814,
"self": 194.55047395990186,
"children": {
"RLTrainer._checkpoint": {
"total": 0.27703473400015355,
"count": 3,
"self": 0.27703473400015355
}
}
},
"_update_policy": {
"total": 743.0071916920033,
"count": 648,
"self": 282.6112319938966,
"children": {
"TorchPPOOptimizer.update": {
"total": 460.3959596981067,
"count": 34233,
"self": 460.3959596981067
}
}
}
}
}
}
},
"trainer_threads": {
"total": 9.509994924883358e-07,
"count": 1,
"self": 9.509994924883358e-07
},
"TrainerController._save_models": {
"total": 0.08724590500060003,
"count": 1,
"self": 0.0012740770007440005,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08597182799985603,
"count": 1,
"self": 0.08597182799985603
}
}
}
}
}
}
}