ppo-Pyramids / run_logs /timers.json
sergey-antonov's picture
v1
ed50233
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.33395376801490784,
"min": 0.31589969992637634,
"max": 1.441728949546814,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 10082.732421875,
"min": 9608.4052734375,
"max": 43736.2890625,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989998.0,
"min": 29984.0,
"max": 989998.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989998.0,
"min": 29984.0,
"max": 989998.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.4393922686576843,
"min": -0.09438030421733856,
"max": 0.5326537489891052,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 118.63591003417969,
"min": -22.93441390991211,
"max": 145.4144744873047,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.006035422440618277,
"min": -0.043881602585315704,
"max": 0.37190476059913635,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 1.6295640468597412,
"min": -12.286849021911621,
"max": 88.51333618164062,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06893740853928819,
"min": 0.06297610218352011,
"max": 0.0722759968534685,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9651237195500346,
"min": 0.5037696469214212,
"max": 1.0496780267664387,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.014959285859808035,
"min": 5.406075954345274e-05,
"max": 0.014959285859808035,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.2094300020373125,
"min": 0.0007027898740648856,
"max": 0.2094300020373125,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.5040617843928566e-06,
"min": 7.5040617843928566e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00010505686498149999,
"min": 0.00010505686498149999,
"max": 0.0035067146310951993,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10250132142857142,
"min": 0.10250132142857142,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4350185,
"min": 1.3886848,
"max": 2.5689048,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0002598820107142857,
"min": 0.0002598820107142857,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.00363834815,
"min": 0.00363834815,
"max": 0.11691358952,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.008380013518035412,
"min": 0.008380013518035412,
"max": 0.5020480155944824,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.11732018738985062,
"min": 0.11732018738985062,
"max": 3.514336109161377,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 452.57575757575756,
"min": 356.89285714285717,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 29870.0,
"min": 16783.0,
"max": 33037.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.4333671508217924,
"min": -0.9998500514775515,
"max": 1.5676048004842666,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 96.0355991050601,
"min": -31.995201647281647,
"max": 130.11119844019413,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.4333671508217924,
"min": -0.9998500514775515,
"max": 1.5676048004842666,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 96.0355991050601,
"min": -31.995201647281647,
"max": 130.11119844019413,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.038773944413327655,
"min": 0.03263499954342281,
"max": 9.759006653638448,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 2.597854275692953,
"min": 2.597854275692953,
"max": 165.9031131118536,
"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": "1676500766",
"python_version": "3.9.7 (default, Sep 16 2021, 13:09:58) \n[GCC 7.5.0]",
"command_line_arguments": "/home/hit/app/hf39/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",
"numpy_version": "1.23.5",
"end_time_seconds": "1676502001"
},
"total": 1234.8806504129989,
"count": 1,
"self": 0.27023988099790586,
"children": {
"run_training.setup": {
"total": 0.00818264299959992,
"count": 1,
"self": 0.00818264299959992
},
"TrainerController.start_learning": {
"total": 1234.6022278890014,
"count": 1,
"self": 0.9804137439532497,
"children": {
"TrainerController._reset_env": {
"total": 0.5893832139990991,
"count": 1,
"self": 0.5893832139990991
},
"TrainerController.advance": {
"total": 1232.9404128950482,
"count": 63639,
"self": 0.8615264128366107,
"children": {
"env_step": {
"total": 665.5740389391613,
"count": 63639,
"self": 607.0765119097432,
"children": {
"SubprocessEnvManager._take_step": {
"total": 57.89052784738669,
"count": 63639,
"self": 2.5971530292208627,
"children": {
"TorchPolicy.evaluate": {
"total": 55.29337481816583,
"count": 62546,
"self": 10.42145294416332,
"children": {
"TorchPolicy.sample_actions": {
"total": 44.87192187400251,
"count": 62546,
"self": 44.87192187400251
}
}
}
}
},
"workers": {
"total": 0.6069991820313589,
"count": 63639,
"self": 0.0,
"children": {
"worker_root": {
"total": 1233.1645771758867,
"count": 63639,
"is_parallel": true,
"self": 686.3652863221578,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.00158286300029431,
"count": 1,
"is_parallel": true,
"self": 0.000465986999188317,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001116876001105993,
"count": 8,
"is_parallel": true,
"self": 0.001116876001105993
}
}
},
"UnityEnvironment.step": {
"total": 0.021419792999950005,
"count": 1,
"is_parallel": true,
"self": 0.0003284169997641584,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00023260899979504757,
"count": 1,
"is_parallel": true,
"self": 0.00023260899979504757
},
"communicator.exchange": {
"total": 0.020013343000755413,
"count": 1,
"is_parallel": true,
"self": 0.020013343000755413
},
"steps_from_proto": {
"total": 0.0008454239996353863,
"count": 1,
"is_parallel": true,
"self": 0.00020051000319654122,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0006449139964388451,
"count": 8,
"is_parallel": true,
"self": 0.0006449139964388451
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 546.799290853729,
"count": 63638,
"is_parallel": true,
"self": 20.61271538584697,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 12.430857172174,
"count": 63638,
"is_parallel": true,
"self": 12.430857172174
},
"communicator.exchange": {
"total": 464.3190750559097,
"count": 63638,
"is_parallel": true,
"self": 464.3190750559097
},
"steps_from_proto": {
"total": 49.436643239798286,
"count": 63638,
"is_parallel": true,
"self": 11.509283125888032,
"children": {
"_process_rank_one_or_two_observation": {
"total": 37.927360113910254,
"count": 509104,
"is_parallel": true,
"self": 37.927360113910254
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 566.5048475430503,
"count": 63639,
"self": 1.7729522932768305,
"children": {
"process_trajectory": {
"total": 97.57101056479951,
"count": 63639,
"self": 97.41618787679909,
"children": {
"RLTrainer._checkpoint": {
"total": 0.1548226880004222,
"count": 2,
"self": 0.1548226880004222
}
}
},
"_update_policy": {
"total": 467.160884684974,
"count": 453,
"self": 132.26211793892617,
"children": {
"TorchPPOOptimizer.update": {
"total": 334.89876674604784,
"count": 22767,
"self": 334.89876674604784
}
}
}
}
}
}
},
"trainer_threads": {
"total": 5.9100057114847e-07,
"count": 1,
"self": 5.9100057114847e-07
},
"TrainerController._save_models": {
"total": 0.09201744500023779,
"count": 1,
"self": 0.00161428200044611,
"children": {
"RLTrainer._checkpoint": {
"total": 0.09040316299979168,
"count": 1,
"self": 0.09040316299979168
}
}
}
}
}
}
}