QuickSilver007's picture
First Pyramids on GPU
9aab330
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.7325840592384338,
"min": 0.7325840592384338,
"max": 1.500280499458313,
"count": 16
},
"Pyramids.Policy.Entropy.sum": {
"value": 21766.537109375,
"min": 21766.537109375,
"max": 45512.5078125,
"count": 16
},
"Pyramids.Step.mean": {
"value": 479901.0,
"min": 29952.0,
"max": 479901.0,
"count": 16
},
"Pyramids.Step.sum": {
"value": 479901.0,
"min": 29952.0,
"max": 479901.0,
"count": 16
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.011484791524708271,
"min": -0.1566089391708374,
"max": 0.07857011258602142,
"count": 16
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 2.8137738704681396,
"min": -37.11631774902344,
"max": 19.3282470703125,
"count": 16
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.02296760492026806,
"min": 0.02070874348282814,
"max": 0.2647741138935089,
"count": 16
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 5.627063274383545,
"min": 5.094350814819336,
"max": 64.07533264160156,
"count": 16
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06339376812782178,
"min": 0.06339376812782178,
"max": 0.07245769744995648,
"count": 16
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.8875127537895049,
"min": 0.4901012743678715,
"max": 1.0313972472407236,
"count": 16
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.004799517112483185,
"min": 0.00025910588923763284,
"max": 0.00933214845968003,
"count": 16
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.06719323957476458,
"min": 0.0031092706708515942,
"max": 0.08684837874837588,
"count": 16
},
"Pyramids.Policy.LearningRate.mean": {
"value": 2.0630278837557143e-05,
"min": 2.0630278837557143e-05,
"max": 0.00029030126037577137,
"count": 16
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.0002888239037258,
"min": 0.0002888239037258,
"max": 0.0030639523786825996,
"count": 16
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10687672857142856,
"min": 0.10687672857142856,
"max": 0.19676708571428575,
"count": 16
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4962741999999998,
"min": 1.3773696000000002,
"max": 2.2213174,
"count": 16
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0006969851842857143,
"min": 0.0006969851842857143,
"max": 0.00967703186285714,
"count": 16
},
"Pyramids.Policy.Beta.sum": {
"value": 0.00975779258,
"min": 0.00975779258,
"max": 0.10214960826000001,
"count": 16
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.022607997059822083,
"min": 0.022607997059822083,
"max": 0.4268878400325775,
"count": 16
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.31651195883750916,
"min": 0.31651195883750916,
"max": 2.9882149696350098,
"count": 16
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 857.6764705882352,
"min": 803.7435897435897,
"max": 999.0,
"count": 16
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 29161.0,
"min": 15984.0,
"max": 33013.0,
"count": 16
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": -0.1521882796112229,
"min": -1.0000000521540642,
"max": 0.06768201826474605,
"count": 16
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": -5.174401506781578,
"min": -31.99600164592266,
"max": 2.639598712325096,
"count": 16
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": -0.1521882796112229,
"min": -1.0000000521540642,
"max": 0.06768201826474605,
"count": 16
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": -5.174401506781578,
"min": -31.99600164592266,
"max": 2.639598712325096,
"count": 16
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.20040673938638065,
"min": 0.19343742634504077,
"max": 8.914730155840516,
"count": 16
},
"Pyramids.Policy.RndReward.sum": {
"value": 6.813829139136942,
"min": 6.813829139136942,
"max": 142.63568249344826,
"count": 16
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 16
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 16
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1658259079",
"python_version": "3.7.13 (default, Apr 24 2022, 01:04:09) \n[GCC 7.5.0]",
"command_line_arguments": "/usr/local/bin/mlagents-learn ./config/ppo/PyramidsRND.yaml --env=./trained-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": "1658260040"
},
"total": 961.211765715,
"count": 1,
"self": 0.43596296399994117,
"children": {
"run_training.setup": {
"total": 0.04359825500000625,
"count": 1,
"self": 0.04359825500000625
},
"TrainerController.start_learning": {
"total": 960.732204496,
"count": 1,
"self": 0.6374291890089125,
"children": {
"TrainerController._reset_env": {
"total": 10.123609308999903,
"count": 1,
"self": 10.123609308999903
},
"TrainerController.advance": {
"total": 949.877687780991,
"count": 31572,
"self": 0.6696866799652526,
"children": {
"env_step": {
"total": 595.5912213900218,
"count": 31572,
"self": 543.862243709017,
"children": {
"SubprocessEnvManager._take_step": {
"total": 51.39484722500151,
"count": 31572,
"self": 2.224013943980026,
"children": {
"TorchPolicy.evaluate": {
"total": 49.17083328102149,
"count": 31305,
"self": 16.94518822401926,
"children": {
"TorchPolicy.sample_actions": {
"total": 32.225645057002225,
"count": 31305,
"self": 32.225645057002225
}
}
}
}
},
"workers": {
"total": 0.33413045600332225,
"count": 31572,
"self": 0.0,
"children": {
"worker_root": {
"total": 958.7775657760201,
"count": 31572,
"is_parallel": true,
"self": 463.5413441930268,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.007156846000043515,
"count": 1,
"is_parallel": true,
"self": 0.0038985030001867926,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0032583429998567226,
"count": 8,
"is_parallel": true,
"self": 0.0032583429998567226
}
}
},
"UnityEnvironment.step": {
"total": 0.04677170100001149,
"count": 1,
"is_parallel": true,
"self": 0.0004995409999537515,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00040436900007989607,
"count": 1,
"is_parallel": true,
"self": 0.00040436900007989607
},
"communicator.exchange": {
"total": 0.04426529499994558,
"count": 1,
"is_parallel": true,
"self": 0.04426529499994558
},
"steps_from_proto": {
"total": 0.0016024960000322608,
"count": 1,
"is_parallel": true,
"self": 0.00041890900013186183,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001183586999900399,
"count": 8,
"is_parallel": true,
"self": 0.001183586999900399
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 495.23622158299327,
"count": 31571,
"is_parallel": true,
"self": 13.317088250992924,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 11.285431233010854,
"count": 31571,
"is_parallel": true,
"self": 11.285431233010854
},
"communicator.exchange": {
"total": 425.64785143300287,
"count": 31571,
"is_parallel": true,
"self": 425.64785143300287
},
"steps_from_proto": {
"total": 44.98585066598662,
"count": 31571,
"is_parallel": true,
"self": 10.914046805989983,
"children": {
"_process_rank_one_or_two_observation": {
"total": 34.071803859996635,
"count": 252568,
"is_parallel": true,
"self": 34.071803859996635
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 353.6167797110039,
"count": 31572,
"self": 1.0527367820033078,
"children": {
"process_trajectory": {
"total": 78.48377242100173,
"count": 31572,
"self": 78.3806933400017,
"children": {
"RLTrainer._checkpoint": {
"total": 0.1030790810000326,
"count": 1,
"self": 0.1030790810000326
}
}
},
"_update_policy": {
"total": 274.0802705079989,
"count": 213,
"self": 110.47030244500138,
"children": {
"TorchPPOOptimizer.update": {
"total": 163.6099680629975,
"count": 11397,
"self": 163.6099680629975
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.1180000001331791e-06,
"count": 1,
"self": 1.1180000001331791e-06
},
"TrainerController._save_models": {
"total": 0.09347709900021073,
"count": 1,
"self": 0.0015976420002061786,
"children": {
"RLTrainer._checkpoint": {
"total": 0.09187945700000455,
"count": 1,
"self": 0.09187945700000455
}
}
}
}
}
}
}