ppo-Pyramids / run_logs /timers.json
mingdinghan's picture
First Push
0d01434
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.5739387273788452,
"min": 0.5739387273788452,
"max": 1.4350367784500122,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 17236.52734375,
"min": 17236.52734375,
"max": 43533.27734375,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989886.0,
"min": 29952.0,
"max": 989886.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989886.0,
"min": 29952.0,
"max": 989886.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.38323527574539185,
"min": -0.1159958764910698,
"max": 0.4755720794200897,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 101.17411041259766,
"min": -28.071002960205078,
"max": 129.8311767578125,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": -0.8554849624633789,
"min": -0.8554849624633789,
"max": 0.5400704145431519,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": -225.8480224609375,
"min": -225.8480224609375,
"max": 142.57858276367188,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06845733564786766,
"min": 0.06442001090978185,
"max": 0.07498557920025122,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9584026990701472,
"min": 0.48873475858514814,
"max": 1.0592627647925195,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.11830140795954568,
"min": 7.568561299922972e-05,
"max": 0.11830140795954568,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 1.6562197114336394,
"min": 0.0009839129689899863,
"max": 1.6562197114336394,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.720468855114278e-06,
"min": 7.720468855114278e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.0001080865639715999,
"min": 0.0001080865639715999,
"max": 0.0033840098719968,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10257345714285715,
"min": 0.10257345714285715,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4360284,
"min": 1.3886848,
"max": 2.5280032,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00026708836857142844,
"min": 0.00026708836857142844,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.003739237159999998,
"min": 0.003739237159999998,
"max": 0.11282751968,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.011036752723157406,
"min": 0.011036752723157406,
"max": 0.37027639150619507,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.15451453626155853,
"min": 0.15451453626155853,
"max": 2.5919346809387207,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 452.7014925373134,
"min": 411.0845070422535,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 30331.0,
"min": 15984.0,
"max": 33017.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.3382746024585481,
"min": -1.0000000521540642,
"max": 1.5270337731065884,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 89.66439836472273,
"min": -30.647001668810844,
"max": 113.12339881062508,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.3382746024585481,
"min": -1.0000000521540642,
"max": 1.5270337731065884,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 89.66439836472273,
"min": -30.647001668810844,
"max": 113.12339881062508,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.051919059223259116,
"min": 0.04706497416133061,
"max": 7.091442469507456,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 3.478576967958361,
"min": 3.202025212798617,
"max": 113.4630795121193,
"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": "1677374220",
"python_version": "3.8.10 (default, Nov 14 2022, 12:59:47) \n[GCC 9.4.0]",
"command_line_arguments": "/usr/local/bin/mlagents-learn ./config/ppo/PyramidsRND.yaml --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids Training 3 --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": "1.22.4",
"end_time_seconds": "1677376393"
},
"total": 2173.434751767,
"count": 1,
"self": 0.43658382100011295,
"children": {
"run_training.setup": {
"total": 0.1042658730000312,
"count": 1,
"self": 0.1042658730000312
},
"TrainerController.start_learning": {
"total": 2172.893902073,
"count": 1,
"self": 1.3490137349108409,
"children": {
"TrainerController._reset_env": {
"total": 8.857764659000054,
"count": 1,
"self": 8.857764659000054
},
"TrainerController.advance": {
"total": 2162.6012864380887,
"count": 63448,
"self": 1.346102698107643,
"children": {
"env_step": {
"total": 1432.5092960480015,
"count": 63448,
"self": 1322.848226275944,
"children": {
"SubprocessEnvManager._take_step": {
"total": 108.8803106070618,
"count": 63448,
"self": 4.5962412241109405,
"children": {
"TorchPolicy.evaluate": {
"total": 104.28406938295086,
"count": 62570,
"self": 35.559176850931294,
"children": {
"TorchPolicy.sample_actions": {
"total": 68.72489253201957,
"count": 62570,
"self": 68.72489253201957
}
}
}
}
},
"workers": {
"total": 0.7807591649958567,
"count": 63448,
"self": 0.0,
"children": {
"worker_root": {
"total": 2168.3562094329614,
"count": 63448,
"is_parallel": true,
"self": 956.0042450539206,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.003538267000067208,
"count": 1,
"is_parallel": true,
"self": 0.0023917950003351507,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011464719997320572,
"count": 8,
"is_parallel": true,
"self": 0.0011464719997320572
}
}
},
"UnityEnvironment.step": {
"total": 0.046180054999922504,
"count": 1,
"is_parallel": true,
"self": 0.0004983079999192341,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0004652260000739261,
"count": 1,
"is_parallel": true,
"self": 0.0004652260000739261
},
"communicator.exchange": {
"total": 0.0436189329999479,
"count": 1,
"is_parallel": true,
"self": 0.0436189329999479
},
"steps_from_proto": {
"total": 0.0015975879999814424,
"count": 1,
"is_parallel": true,
"self": 0.00040437299992390763,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011932150000575348,
"count": 8,
"is_parallel": true,
"self": 0.0011932150000575348
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1212.3519643790407,
"count": 63447,
"is_parallel": true,
"self": 30.817458397094697,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 22.563131756993926,
"count": 63447,
"is_parallel": true,
"self": 22.563131756993926
},
"communicator.exchange": {
"total": 1070.1979765269828,
"count": 63447,
"is_parallel": true,
"self": 1070.1979765269828
},
"steps_from_proto": {
"total": 88.77339769796936,
"count": 63447,
"is_parallel": true,
"self": 20.732881078994865,
"children": {
"_process_rank_one_or_two_observation": {
"total": 68.0405166189745,
"count": 507576,
"is_parallel": true,
"self": 68.0405166189745
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 728.7458876919795,
"count": 63448,
"self": 2.439960034019691,
"children": {
"process_trajectory": {
"total": 159.58618741896646,
"count": 63448,
"self": 159.40425509396653,
"children": {
"RLTrainer._checkpoint": {
"total": 0.18193232499993428,
"count": 2,
"self": 0.18193232499993428
}
}
},
"_update_policy": {
"total": 566.7197402389934,
"count": 448,
"self": 221.5583025919941,
"children": {
"TorchPPOOptimizer.update": {
"total": 345.1614376469993,
"count": 22785,
"self": 345.1614376469993
}
}
}
}
}
}
},
"trainer_threads": {
"total": 8.199999683711212e-07,
"count": 1,
"self": 8.199999683711212e-07
},
"TrainerController._save_models": {
"total": 0.08583642100029465,
"count": 1,
"self": 0.0014336260005620716,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08440279499973258,
"count": 1,
"self": 0.08440279499973258
}
}
}
}
}
}
}