ppo-PyramidsRND / run_logs /timers.json
SergejSchweizer's picture
First Push
fbb36f6
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.3972884714603424,
"min": 0.3972884714603424,
"max": 1.459014892578125,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 12058.5,
"min": 12058.5,
"max": 44260.67578125,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989936.0,
"min": 29938.0,
"max": 989936.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989936.0,
"min": 29938.0,
"max": 989936.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.4479498267173767,
"min": -0.15801416337490082,
"max": 0.47586992383003235,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 119.15465545654297,
"min": -37.44935607910156,
"max": 128.48487854003906,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 7.693795204162598,
"min": -0.7297115325927734,
"max": 7.693795204162598,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 2046.549560546875,
"min": -196.2924041748047,
"max": 2046.549560546875,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.07124909386145305,
"min": 0.06443510237839581,
"max": 0.07235805791189975,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 1.0687364079217958,
"min": 0.4904050968999535,
"max": 1.0687364079217958,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 3.967677446347631,
"min": 0.00017520919672265258,
"max": 3.967677446347631,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 59.515161695214466,
"min": 0.0022777195573944835,
"max": 59.515161695214466,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.4704175098933355e-06,
"min": 7.4704175098933355e-06,
"max": 0.0002952335158745429,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00011205626264840003,
"min": 0.00011205626264840003,
"max": 0.0033715954761348995,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10249010666666668,
"min": 0.10249010666666668,
"max": 0.19841117142857143,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.5373516000000003,
"min": 1.3888782,
"max": 2.4434698,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0002587616560000001,
"min": 0.0002587616560000001,
"max": 0.009841276025714286,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.003881424840000001,
"min": 0.003881424840000001,
"max": 0.11240412349000001,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.020568164065480232,
"min": 0.020568164065480232,
"max": 0.439841091632843,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.30852246284484863,
"min": 0.3011903166770935,
"max": 3.078887701034546,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 412.7432432432432,
"min": 380.1518987341772,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 30543.0,
"min": 16609.0,
"max": 33599.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.4521026792558465,
"min": -0.9998125522397459,
"max": 1.5691974412979959,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 107.45559826493263,
"min": -31.994001671671867,
"max": 123.96659786254168,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.4521026792558465,
"min": -0.9998125522397459,
"max": 1.5691974412979959,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 107.45559826493263,
"min": -31.994001671671867,
"max": 123.96659786254168,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.08748003310599439,
"min": 0.08418644770276093,
"max": 8.21735037424985,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 6.4735224498435855,
"min": 6.434110916714417,
"max": 139.69495636224747,
"count": 33
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 33
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 33
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1673643521",
"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=Pyramids Training --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": "1673645872"
},
"total": 2351.272789075,
"count": 1,
"self": 0.5044134880004094,
"children": {
"run_training.setup": {
"total": 0.12302808699996604,
"count": 1,
"self": 0.12302808699996604
},
"TrainerController.start_learning": {
"total": 2350.6453475,
"count": 1,
"self": 1.505874720915017,
"children": {
"TrainerController._reset_env": {
"total": 7.063458406999871,
"count": 1,
"self": 7.063458406999871
},
"TrainerController.advance": {
"total": 2341.9741817410845,
"count": 63620,
"self": 1.6278781102278117,
"children": {
"env_step": {
"total": 1580.929021931935,
"count": 63620,
"self": 1457.5443916889553,
"children": {
"SubprocessEnvManager._take_step": {
"total": 122.35888034404024,
"count": 63620,
"self": 4.900491438964764,
"children": {
"TorchPolicy.evaluate": {
"total": 117.45838890507548,
"count": 62560,
"self": 39.083532314174136,
"children": {
"TorchPolicy.sample_actions": {
"total": 78.37485659090135,
"count": 62560,
"self": 78.37485659090135
}
}
}
}
},
"workers": {
"total": 1.0257498989394662,
"count": 63620,
"self": 0.0,
"children": {
"worker_root": {
"total": 2345.298817686168,
"count": 63620,
"is_parallel": true,
"self": 1004.8611989801025,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.00193316299964863,
"count": 1,
"is_parallel": true,
"self": 0.0006355640007313923,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0012975989989172376,
"count": 8,
"is_parallel": true,
"self": 0.0012975989989172376
}
}
},
"UnityEnvironment.step": {
"total": 0.05015337500026362,
"count": 1,
"is_parallel": true,
"self": 0.0005645960000038031,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.000507064000430546,
"count": 1,
"is_parallel": true,
"self": 0.000507064000430546
},
"communicator.exchange": {
"total": 0.047222394999607786,
"count": 1,
"is_parallel": true,
"self": 0.047222394999607786
},
"steps_from_proto": {
"total": 0.0018593200002214871,
"count": 1,
"is_parallel": true,
"self": 0.00045896700021330616,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001400353000008181,
"count": 8,
"is_parallel": true,
"self": 0.001400353000008181
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1340.4376187060657,
"count": 63619,
"is_parallel": true,
"self": 33.21204566110191,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 25.99006082293272,
"count": 63619,
"is_parallel": true,
"self": 25.99006082293272
},
"communicator.exchange": {
"total": 1164.1698389980525,
"count": 63619,
"is_parallel": true,
"self": 1164.1698389980525
},
"steps_from_proto": {
"total": 117.06567322397859,
"count": 63619,
"is_parallel": true,
"self": 26.464246274249035,
"children": {
"_process_rank_one_or_two_observation": {
"total": 90.60142694972956,
"count": 508952,
"is_parallel": true,
"self": 90.60142694972956
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 759.4172816989217,
"count": 63620,
"self": 2.9109808408529716,
"children": {
"process_trajectory": {
"total": 164.5548685070694,
"count": 63620,
"self": 164.34181580006907,
"children": {
"RLTrainer._checkpoint": {
"total": 0.21305270700031542,
"count": 2,
"self": 0.21305270700031542
}
}
},
"_update_policy": {
"total": 591.9514323509993,
"count": 450,
"self": 231.5070179009913,
"children": {
"TorchPPOOptimizer.update": {
"total": 360.444414450008,
"count": 22821,
"self": 360.444414450008
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.151000105892308e-06,
"count": 1,
"self": 1.151000105892308e-06
},
"TrainerController._save_models": {
"total": 0.10183148000032816,
"count": 1,
"self": 0.0015592409999953816,
"children": {
"RLTrainer._checkpoint": {
"total": 0.10027223900033277,
"count": 1,
"self": 0.10027223900033277
}
}
}
}
}
}
}