ppo-Pyramids / run_logs /timers.json
odahl's picture
First push
4a16b00
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.47724100947380066,
"min": 0.475818932056427,
"max": 1.4745895862579346,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 14385.953125,
"min": 14305.0205078125,
"max": 44733.1484375,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989928.0,
"min": 29882.0,
"max": 989928.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989928.0,
"min": 29882.0,
"max": 989928.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.509105920791626,
"min": -0.11253941804170609,
"max": 0.509105920791626,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 137.9676971435547,
"min": -27.234539031982422,
"max": 137.9676971435547,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.08279137313365936,
"min": -0.02762259542942047,
"max": 0.33408859372138977,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 22.43646240234375,
"min": -7.264742374420166,
"max": 79.17899322509766,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06777677917852996,
"min": 0.0628374158305309,
"max": 0.072562707381578,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9488749084994196,
"min": 0.4937263054289151,
"max": 1.0376452305160153,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.014845192848865518,
"min": 0.0007471472953801013,
"max": 0.017985347336174788,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.20783269988411726,
"min": 0.008218620249181114,
"max": 0.251794862706447,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.4055332458071435e-06,
"min": 7.4055332458071435e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00010367746544130001,
"min": 0.00010367746544130001,
"max": 0.003377716674094499,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10246847857142857,
"min": 0.10246847857142857,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4345587,
"min": 1.3886848,
"max": 2.4259055000000003,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00025660100928571425,
"min": 0.00025660100928571425,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.00359241413,
"min": 0.00359241413,
"max": 0.11260795944999999,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.008719226345419884,
"min": 0.008421015925705433,
"max": 0.4308992028236389,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.12206916511058807,
"min": 0.11789421737194061,
"max": 3.016294479370117,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 376.5375,
"min": 370.43037974683546,
"max": 989.375,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 30123.0,
"min": 16809.0,
"max": 32170.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.4984324678778649,
"min": -0.9277750519104302,
"max": 1.5789265587360044,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 119.87459743022919,
"min": -29.688801661133766,
"max": 124.73519814014435,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.4984324678778649,
"min": -0.9277750519104302,
"max": 1.5789265587360044,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 119.87459743022919,
"min": -29.688801661133766,
"max": 124.73519814014435,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.03403511967480881,
"min": 0.03292561192017585,
"max": 8.617009348729077,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 2.722809573984705,
"min": 2.601123341693892,
"max": 146.48915892839432,
"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": "1675188509",
"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 --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": "1675190433"
},
"total": 1924.0596803820001,
"count": 1,
"self": 0.4743553790003716,
"children": {
"run_training.setup": {
"total": 0.09756109899990406,
"count": 1,
"self": 0.09756109899990406
},
"TrainerController.start_learning": {
"total": 1923.4877639039998,
"count": 1,
"self": 1.2060693920232097,
"children": {
"TrainerController._reset_env": {
"total": 5.82987182099987,
"count": 1,
"self": 5.82987182099987
},
"TrainerController.advance": {
"total": 1916.3669391769763,
"count": 63546,
"self": 1.243221094866385,
"children": {
"env_step": {
"total": 1279.4120324430642,
"count": 63546,
"self": 1174.6431464429943,
"children": {
"SubprocessEnvManager._take_step": {
"total": 104.0189159700501,
"count": 63546,
"self": 4.188172177093293,
"children": {
"TorchPolicy.evaluate": {
"total": 99.83074379295681,
"count": 62562,
"self": 33.45478156292347,
"children": {
"TorchPolicy.sample_actions": {
"total": 66.37596223003334,
"count": 62562,
"self": 66.37596223003334
}
}
}
}
},
"workers": {
"total": 0.7499700300197674,
"count": 63546,
"self": 0.0,
"children": {
"worker_root": {
"total": 1919.9521882260833,
"count": 63546,
"is_parallel": true,
"self": 842.5860062051206,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0016193189999285096,
"count": 1,
"is_parallel": true,
"self": 0.0005616170001303544,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0010577019997981552,
"count": 8,
"is_parallel": true,
"self": 0.0010577019997981552
}
}
},
"UnityEnvironment.step": {
"total": 0.045508490999964124,
"count": 1,
"is_parallel": true,
"self": 0.000471306999997978,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00042623500007721304,
"count": 1,
"is_parallel": true,
"self": 0.00042623500007721304
},
"communicator.exchange": {
"total": 0.043040635999886945,
"count": 1,
"is_parallel": true,
"self": 0.043040635999886945
},
"steps_from_proto": {
"total": 0.0015703130000019883,
"count": 1,
"is_parallel": true,
"self": 0.0004356219999408495,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011346910000611388,
"count": 8,
"is_parallel": true,
"self": 0.0011346910000611388
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1077.3661820209627,
"count": 63545,
"is_parallel": true,
"self": 26.71598467695594,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 22.422388595976372,
"count": 63545,
"is_parallel": true,
"self": 22.422388595976372
},
"communicator.exchange": {
"total": 936.6587674520763,
"count": 63545,
"is_parallel": true,
"self": 936.6587674520763
},
"steps_from_proto": {
"total": 91.5690412959541,
"count": 63545,
"is_parallel": true,
"self": 21.64290376375243,
"children": {
"_process_rank_one_or_two_observation": {
"total": 69.92613753220166,
"count": 508360,
"is_parallel": true,
"self": 69.92613753220166
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 635.7116856390458,
"count": 63546,
"self": 2.257667536023291,
"children": {
"process_trajectory": {
"total": 142.88804267502087,
"count": 63546,
"self": 142.71259332202112,
"children": {
"RLTrainer._checkpoint": {
"total": 0.17544935299974895,
"count": 2,
"self": 0.17544935299974895
}
}
},
"_update_policy": {
"total": 490.5659754280016,
"count": 450,
"self": 182.74471138298827,
"children": {
"TorchPPOOptimizer.update": {
"total": 307.8212640450133,
"count": 22773,
"self": 307.8212640450133
}
}
}
}
}
}
},
"trainer_threads": {
"total": 9.970003702619579e-07,
"count": 1,
"self": 9.970003702619579e-07
},
"TrainerController._save_models": {
"total": 0.08488251700009641,
"count": 1,
"self": 0.0014942160000828153,
"children": {
"RLTrainer._checkpoint": {
"total": 0.0833883010000136,
"count": 1,
"self": 0.0833883010000136
}
}
}
}
}
}
}