ppo-Pyramid / run_logs /timers.json
jvogie's picture
First Push
eabaf62
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.37749379873275757,
"min": 0.3668733835220337,
"max": 1.521689534187317,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 11270.455078125,
"min": 10947.501953125,
"max": 46161.97265625,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989957.0,
"min": 29952.0,
"max": 989957.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989957.0,
"min": 29952.0,
"max": 989957.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.41523662209510803,
"min": -0.07750914245843887,
"max": 0.5447093844413757,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 109.62246704101562,
"min": -18.679702758789062,
"max": 148.16094970703125,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.01273332443088293,
"min": -0.007094942964613438,
"max": 0.3513241112232208,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 3.3615975379943848,
"min": -1.78083074092865,
"max": 83.2638168334961,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06564585840983671,
"min": 0.06375341781732008,
"max": 0.07398141622166697,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9190420177377139,
"min": 0.49288579604177657,
"max": 1.092725614105196,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.012811746558457213,
"min": 0.0009522398337926697,
"max": 0.015005599704593241,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.179364451818401,
"min": 0.012379117839304706,
"max": 0.21401302119421398,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.582833186707145e-06,
"min": 7.582833186707145e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00010615966461390003,
"min": 0.00010615966461390003,
"max": 0.0035097740300753994,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10252757857142858,
"min": 0.10252757857142858,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4353861,
"min": 1.3886848,
"max": 2.5699246000000002,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00026250509928571437,
"min": 0.00026250509928571437,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0036750713900000013,
"min": 0.0036750713900000013,
"max": 0.11701546754,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.009747141040861607,
"min": 0.009747141040861607,
"max": 0.3306867480278015,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.13645997643470764,
"min": 0.13645997643470764,
"max": 2.314807176589966,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 421.69117647058823,
"min": 343.1,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 28675.0,
"min": 15984.0,
"max": 33049.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.4017940905383404,
"min": -1.0000000521540642,
"max": 1.5818774787709118,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 95.32199815660715,
"min": -29.94700165838003,
"max": 134.2333985492587,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.4017940905383404,
"min": -1.0000000521540642,
"max": 1.5818774787709118,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 95.32199815660715,
"min": -29.94700165838003,
"max": 134.2333985492587,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.04215115118901903,
"min": 0.03686977436094215,
"max": 6.094980431720614,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 2.866278280853294,
"min": 2.866278280853294,
"max": 97.51968690752983,
"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": "1698091267",
"python_version": "3.10.12 (main, Jun 11 2023, 05:26:28) [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",
"mlagents_version": "1.1.0.dev0",
"mlagents_envs_version": "1.1.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.1.0+cu118",
"numpy_version": "1.23.5",
"end_time_seconds": "1698093823"
},
"total": 2555.796335392,
"count": 1,
"self": 0.5429832220002027,
"children": {
"run_training.setup": {
"total": 0.047860105999916414,
"count": 1,
"self": 0.047860105999916414
},
"TrainerController.start_learning": {
"total": 2555.205492064,
"count": 1,
"self": 1.7626626940923416,
"children": {
"TrainerController._reset_env": {
"total": 4.810568804000013,
"count": 1,
"self": 4.810568804000013
},
"TrainerController.advance": {
"total": 2548.547839016908,
"count": 63957,
"self": 1.8301807611042022,
"children": {
"env_step": {
"total": 1842.248575575849,
"count": 63957,
"self": 1683.6018749669834,
"children": {
"SubprocessEnvManager._take_step": {
"total": 157.56542295289046,
"count": 63957,
"self": 5.581641860741456,
"children": {
"TorchPolicy.evaluate": {
"total": 151.983781092149,
"count": 62552,
"self": 151.983781092149
}
}
},
"workers": {
"total": 1.0812776559751,
"count": 63957,
"self": 0.0,
"children": {
"worker_root": {
"total": 2549.3279822340555,
"count": 63957,
"is_parallel": true,
"self": 1002.4033721949736,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0020522640002127446,
"count": 1,
"is_parallel": true,
"self": 0.0006544660009240033,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0013977979992887413,
"count": 8,
"is_parallel": true,
"self": 0.0013977979992887413
}
}
},
"UnityEnvironment.step": {
"total": 0.04945132100010596,
"count": 1,
"is_parallel": true,
"self": 0.0006574020003426995,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0005272529997455422,
"count": 1,
"is_parallel": true,
"self": 0.0005272529997455422
},
"communicator.exchange": {
"total": 0.04648980299998584,
"count": 1,
"is_parallel": true,
"self": 0.04648980299998584
},
"steps_from_proto": {
"total": 0.0017768630000318808,
"count": 1,
"is_parallel": true,
"self": 0.00037249800016070367,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0014043649998711771,
"count": 8,
"is_parallel": true,
"self": 0.0014043649998711771
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1546.924610039082,
"count": 63956,
"is_parallel": true,
"self": 39.49725398894225,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 26.229093499036026,
"count": 63956,
"is_parallel": true,
"self": 26.229093499036026
},
"communicator.exchange": {
"total": 1369.5760900640003,
"count": 63956,
"is_parallel": true,
"self": 1369.5760900640003
},
"steps_from_proto": {
"total": 111.6221724871034,
"count": 63956,
"is_parallel": true,
"self": 23.949024452989306,
"children": {
"_process_rank_one_or_two_observation": {
"total": 87.67314803411409,
"count": 511648,
"is_parallel": true,
"self": 87.67314803411409
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 704.4690826799547,
"count": 63957,
"self": 3.3797538389248984,
"children": {
"process_trajectory": {
"total": 135.49602493103203,
"count": 63957,
"self": 135.26755688203139,
"children": {
"RLTrainer._checkpoint": {
"total": 0.22846804900063944,
"count": 2,
"self": 0.22846804900063944
}
}
},
"_update_policy": {
"total": 565.5933039099978,
"count": 455,
"self": 339.87712636596643,
"children": {
"TorchPPOOptimizer.update": {
"total": 225.71617754403132,
"count": 22800,
"self": 225.71617754403132
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.0989997463184409e-06,
"count": 1,
"self": 1.0989997463184409e-06
},
"TrainerController._save_models": {
"total": 0.08442044999992504,
"count": 1,
"self": 0.0016231130002779537,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08279733699964709,
"count": 1,
"self": 0.08279733699964709
}
}
}
}
}
}
}