ppo-Pyramids / run_logs /timers.json
Croolch's picture
First Push
b87779a verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.27660733461380005,
"min": 0.27660733461380005,
"max": 0.6970478296279907,
"count": 21
},
"Pyramids.Policy.Entropy.sum": {
"value": 8227.408203125,
"min": 8227.408203125,
"max": 21101.03125,
"count": 21
},
"Pyramids.Step.mean": {
"value": 1619896.0,
"min": 1019972.0,
"max": 1619896.0,
"count": 21
},
"Pyramids.Step.sum": {
"value": 1619896.0,
"min": 1019972.0,
"max": 1619896.0,
"count": 21
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.592436671257019,
"min": 0.17887116968631744,
"max": 0.5980270504951477,
"count": 21
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 169.4368896484375,
"min": 38.88249206542969,
"max": 169.4368896484375,
"count": 21
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.010919425636529922,
"min": -0.01185104250907898,
"max": 0.07567622512578964,
"count": 21
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 3.122955799102783,
"min": -3.176079273223877,
"max": 20.810962677001953,
"count": 21
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 304.00943396226415,
"min": 304.00943396226415,
"max": 704.9761904761905,
"count": 21
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 32225.0,
"min": 15375.0,
"max": 33061.0,
"count": 21
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.6582433830876395,
"min": 0.46615810692310333,
"max": 1.6687587479647903,
"count": 21
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 175.7737986072898,
"min": 20.044798597693443,
"max": 175.7737986072898,
"count": 21
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.6582433830876395,
"min": 0.46615810692310333,
"max": 1.6687587479647903,
"count": 21
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 175.7737986072898,
"min": 20.044798597693443,
"max": 175.7737986072898,
"count": 21
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.02843567615113812,
"min": 0.02843567615113812,
"max": 0.09740542004366817,
"count": 21
},
"Pyramids.Policy.RndReward.sum": {
"value": 3.0141816720206407,
"min": 2.240261279977858,
"max": 4.188433061877731,
"count": 21
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06440481870612573,
"min": 0.06440481870612573,
"max": 0.07214185008771558,
"count": 21
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9016674618857602,
"min": 0.6178666375965501,
"max": 1.0821277513157337,
"count": 21
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.0158456465358525,
"min": 0.007521672538129859,
"max": 0.016608913912759385,
"count": 21
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.221839051501935,
"min": 0.0910664204234005,
"max": 0.24370129945015712,
"count": 21
},
"Pyramids.Policy.LearningRate.mean": {
"value": 0.00013951353920979047,
"min": 0.00013951353920979047,
"max": 0.00019892394480314075,
"count": 21
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.0019531895489370665,
"min": 0.0017903155032282668,
"max": 0.0027675777774742666,
"count": 21
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.14650449523809522,
"min": 0.14650449523809522,
"max": 0.16630797037037037,
"count": 21
},
"Pyramids.Policy.Epsilon.sum": {
"value": 2.0510629333333332,
"min": 1.4967717333333335,
"max": 2.422525733333334,
"count": 21
},
"Pyramids.Policy.Beta.mean": {
"value": 0.004655799074285713,
"min": 0.004655799074285713,
"max": 0.006634166239999999,
"count": 21
},
"Pyramids.Policy.Beta.sum": {
"value": 0.06518118703999999,
"min": 0.05970749616,
"max": 0.09231032076000001,
"count": 21
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.008971639908850193,
"min": 0.008432705886662006,
"max": 0.01435074396431446,
"count": 21
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.12560296058654785,
"min": 0.11805787682533264,
"max": 0.1869449019432068,
"count": 21
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 21
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 21
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1710476387",
"python_version": "3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]",
"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": "1.1.0.dev0",
"mlagents_envs_version": "1.1.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.2.1+cu121",
"numpy_version": "1.23.5",
"end_time_seconds": "1710477946"
},
"total": 1558.879617263,
"count": 1,
"self": 0.6296550729998671,
"children": {
"run_training.setup": {
"total": 0.08422484399989116,
"count": 1,
"self": 0.08422484399989116
},
"TrainerController.start_learning": {
"total": 1558.1657373460002,
"count": 1,
"self": 1.012262123995697,
"children": {
"TrainerController._reset_env": {
"total": 2.2892777010001737,
"count": 1,
"self": 2.2892777010001737
},
"TrainerController.advance": {
"total": 1554.8604768780046,
"count": 40547,
"self": 1.0200760621187328,
"children": {
"env_step": {
"total": 1142.0261885059954,
"count": 40547,
"self": 1051.803102436962,
"children": {
"SubprocessEnvManager._take_step": {
"total": 89.596789138031,
"count": 40547,
"self": 3.2497139430465722,
"children": {
"TorchPolicy.evaluate": {
"total": 86.34707519498443,
"count": 39320,
"self": 86.34707519498443
}
}
},
"workers": {
"total": 0.6262969310023436,
"count": 40546,
"self": 0.0,
"children": {
"worker_root": {
"total": 1554.5629688260428,
"count": 40546,
"is_parallel": true,
"self": 587.4225106940371,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.002216930000031425,
"count": 1,
"is_parallel": true,
"self": 0.0006426989998544741,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001574231000176951,
"count": 8,
"is_parallel": true,
"self": 0.001574231000176951
}
}
},
"UnityEnvironment.step": {
"total": 0.05006574899971383,
"count": 1,
"is_parallel": true,
"self": 0.0006695329998365196,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0004964290001225891,
"count": 1,
"is_parallel": true,
"self": 0.0004964290001225891
},
"communicator.exchange": {
"total": 0.04715151199980028,
"count": 1,
"is_parallel": true,
"self": 0.04715151199980028
},
"steps_from_proto": {
"total": 0.0017482749999544467,
"count": 1,
"is_parallel": true,
"self": 0.0003777639994950732,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0013705110004593735,
"count": 8,
"is_parallel": true,
"self": 0.0013705110004593735
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 967.1404581320057,
"count": 40545,
"is_parallel": true,
"self": 23.175993409091006,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 16.720818008942842,
"count": 40545,
"is_parallel": true,
"self": 16.720818008942842
},
"communicator.exchange": {
"total": 856.9368456759912,
"count": 40545,
"is_parallel": true,
"self": 856.9368456759912
},
"steps_from_proto": {
"total": 70.30680103798068,
"count": 40545,
"is_parallel": true,
"self": 14.780989722108643,
"children": {
"_process_rank_one_or_two_observation": {
"total": 55.52581131587203,
"count": 324360,
"is_parallel": true,
"self": 55.52581131587203
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 411.81421230989054,
"count": 40546,
"self": 2.0163928877691433,
"children": {
"process_trajectory": {
"total": 87.82973507513134,
"count": 40546,
"self": 87.69651986713097,
"children": {
"RLTrainer._checkpoint": {
"total": 0.13321520800036524,
"count": 1,
"self": 0.13321520800036524
}
}
},
"_update_policy": {
"total": 321.96808434699005,
"count": 296,
"self": 189.38236025700735,
"children": {
"TorchPPOOptimizer.update": {
"total": 132.5857240899827,
"count": 14277,
"self": 132.5857240899827
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.2219998097862117e-06,
"count": 1,
"self": 1.2219998097862117e-06
},
"TrainerController._save_models": {
"total": 0.0037194209999142913,
"count": 1,
"self": 2.6368999897385947e-05,
"children": {
"RLTrainer._checkpoint": {
"total": 0.0036930520000169054,
"count": 1,
"self": 0.0036930520000169054
}
}
}
}
}
}
}