ppo-Pyramids / run_logs /timers.json
jrauch4's picture
Pyramids - first model
386d10c
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.31022101640701294,
"min": 0.29784926772117615,
"max": 0.7180009484291077,
"count": 100
},
"Pyramids.Policy.Entropy.sum": {
"value": 2958.267578125,
"min": 2780.990234375,
"max": 8822.7958984375,
"count": 100
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 351.44827586206895,
"min": 216.77272727272728,
"max": 548.25,
"count": 100
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 10192.0,
"min": 6728.0,
"max": 13518.0,
"count": 100
},
"Pyramids.Step.mean": {
"value": 1999868.0,
"min": 1009862.0,
"max": 1999868.0,
"count": 100
},
"Pyramids.Step.sum": {
"value": 1999868.0,
"min": 1009862.0,
"max": 1999868.0,
"count": 100
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.7133128046989441,
"min": 0.35823938250541687,
"max": 0.7794756293296814,
"count": 100
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 37.805580139160156,
"min": 17.083505630493164,
"max": 46.76853942871094,
"count": 100
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.0029533114284276962,
"min": -0.014872751198709011,
"max": 0.010451012291014194,
"count": 100
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 0.15652550756931305,
"min": -0.832874059677124,
"max": 0.5852566957473755,
"count": 100
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.5215466310580572,
"min": 1.1238631198280735,
"max": 1.7832272642038085,
"count": 100
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 45.646398931741714,
"min": 21.3533992767334,
"max": 78.46199962496758,
"count": 100
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.5215466310580572,
"min": 1.1238631198280735,
"max": 1.7832272642038085,
"count": 100
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 45.646398931741714,
"min": 21.3533992767334,
"max": 78.46199962496758,
"count": 100
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.019186842233951514,
"min": 0.012917174981504848,
"max": 0.037571362239997026,
"count": 100
},
"Pyramids.Policy.RndReward.sum": {
"value": 0.5756052670185454,
"min": 0.45256442145910114,
"max": 0.9492371222586371,
"count": 100
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.05074516216700431,
"min": 0.04282262219639961,
"max": 0.05563672502175905,
"count": 100
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.05074516216700431,
"min": 0.04282262219639961,
"max": 0.10601836542191449,
"count": 100
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.01610263850307092,
"min": 0.0099980208062334,
"max": 0.019470822933362796,
"count": 100
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.01610263850307092,
"min": 0.0099980208062334,
"max": 0.03894164586672559,
"count": 100
},
"Pyramids.Policy.LearningRate.mean": {
"value": 4.4134991175000417e-07,
"min": 4.4134991175000417e-07,
"max": 0.00024791580041684996,
"count": 100
},
"Pyramids.Policy.LearningRate.sum": {
"value": 4.4134991175000417e-07,
"min": 4.4134991175000417e-07,
"max": 0.00047718735456255,
"count": 100
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10008825000000002,
"min": 0.10008825000000002,
"max": 0.14958315,
"count": 100
},
"Pyramids.Policy.Epsilon.sum": {
"value": 0.10008825000000002,
"min": 0.10008825000000002,
"max": 0.29543745,
"count": 100
},
"Pyramids.Policy.Beta.mean": {
"value": 1.8816175000000088e-05,
"min": 1.8816175000000088e-05,
"max": 0.004963356685000001,
"count": 100
},
"Pyramids.Policy.Beta.sum": {
"value": 1.8816175000000088e-05,
"min": 1.8816175000000088e-05,
"max": 0.009554201254999999,
"count": 100
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.005384649615734816,
"min": 0.005211261101067066,
"max": 0.007747841067612171,
"count": 100
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.005384649615734816,
"min": 0.005211261101067066,
"max": 0.014030024409294128,
"count": 100
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 100
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 100
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1673957392",
"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 --resume",
"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": "1673959797"
},
"total": 2405.385632459999,
"count": 1,
"self": 0.437865072999557,
"children": {
"run_training.setup": {
"total": 0.10557468499973766,
"count": 1,
"self": 0.10557468499973766
},
"TrainerController.start_learning": {
"total": 2404.842192702,
"count": 1,
"self": 1.3430743938906744,
"children": {
"TrainerController._reset_env": {
"total": 6.209119289000228,
"count": 1,
"self": 6.209119289000228
},
"TrainerController.advance": {
"total": 2397.1568622731083,
"count": 64960,
"self": 1.3530111958243651,
"children": {
"env_step": {
"total": 1552.514591648027,
"count": 64960,
"self": 1439.722047392177,
"children": {
"SubprocessEnvManager._take_step": {
"total": 111.94261797900981,
"count": 64960,
"self": 4.311535073033156,
"children": {
"TorchPolicy.evaluate": {
"total": 107.63108290597665,
"count": 62589,
"self": 33.95974611387828,
"children": {
"TorchPolicy.sample_actions": {
"total": 73.67133679209837,
"count": 62589,
"self": 73.67133679209837
}
}
}
}
},
"workers": {
"total": 0.8499262768400513,
"count": 64960,
"self": 0.0,
"children": {
"worker_root": {
"total": 2399.8741675931487,
"count": 64960,
"is_parallel": true,
"self": 1063.525139756065,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0017121780001616571,
"count": 1,
"is_parallel": true,
"self": 0.0006076420004319516,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011045359997297055,
"count": 8,
"is_parallel": true,
"self": 0.0011045359997297055
}
}
},
"UnityEnvironment.step": {
"total": 0.04427539100015565,
"count": 1,
"is_parallel": true,
"self": 0.0005214879993218346,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0004455309999684687,
"count": 1,
"is_parallel": true,
"self": 0.0004455309999684687
},
"communicator.exchange": {
"total": 0.04171734400006244,
"count": 1,
"is_parallel": true,
"self": 0.04171734400006244
},
"steps_from_proto": {
"total": 0.0015910280008029076,
"count": 1,
"is_parallel": true,
"self": 0.0003995300012320513,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011914979995708563,
"count": 8,
"is_parallel": true,
"self": 0.0011914979995708563
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1336.3490278370837,
"count": 64959,
"is_parallel": true,
"self": 27.803562030926514,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 22.657480457138263,
"count": 64959,
"is_parallel": true,
"self": 22.657480457138263
},
"communicator.exchange": {
"total": 1184.3389390060256,
"count": 64959,
"is_parallel": true,
"self": 1184.3389390060256
},
"steps_from_proto": {
"total": 101.54904634299328,
"count": 64959,
"is_parallel": true,
"self": 22.57993668716881,
"children": {
"_process_rank_one_or_two_observation": {
"total": 78.96910965582447,
"count": 519672,
"is_parallel": true,
"self": 78.96910965582447
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 843.2892594292571,
"count": 64960,
"self": 2.8733345402024497,
"children": {
"process_trajectory": {
"total": 133.85696017405553,
"count": 64960,
"self": 133.59503827805474,
"children": {
"RLTrainer._checkpoint": {
"total": 0.26192189600078564,
"count": 2,
"self": 0.26192189600078564
}
}
},
"_update_policy": {
"total": 706.5589647149991,
"count": 120,
"self": 328.8416873281103,
"children": {
"TorchPPOOptimizer.update": {
"total": 377.7172773868888,
"count": 19210,
"self": 377.7172773868888
}
}
}
}
}
}
},
"trainer_threads": {
"total": 9.619998309062794e-07,
"count": 1,
"self": 9.619998309062794e-07
},
"TrainerController._save_models": {
"total": 0.13313578400084225,
"count": 1,
"self": 0.0027590270019572927,
"children": {
"RLTrainer._checkpoint": {
"total": 0.13037675699888496,
"count": 1,
"self": 0.13037675699888496
}
}
}
}
}
}
}