Eric Y
Initial training version
87805fc
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.3468400537967682,
"min": 0.33922168612480164,
"max": 1.4297301769256592,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 10321.9599609375,
"min": 10198.3603515625,
"max": 43372.29296875,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989912.0,
"min": 29905.0,
"max": 989912.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989912.0,
"min": 29905.0,
"max": 989912.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.6016092896461487,
"min": -0.1263759285211563,
"max": 0.6043045520782471,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 169.05221557617188,
"min": -29.951095581054688,
"max": 171.62249755859375,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.02241419069468975,
"min": 0.0013824679190292954,
"max": 0.2965961694717407,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 6.29838752746582,
"min": 0.3718838691711426,
"max": 71.77627563476562,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.07085802641397874,
"min": 0.06606424262558688,
"max": 0.07454688353620534,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9920123697957025,
"min": 0.4932176120665886,
"max": 1.0821457755085375,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.015136433411888512,
"min": 0.0012240963857888107,
"max": 0.01695757400292804,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.21191006776643917,
"min": 0.014689156629465728,
"max": 0.23740603604099253,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 1.509515462701429e-05,
"min": 1.509515462701429e-05,
"max": 0.0005903012587593143,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00021133216477820004,
"min": 0.00021133216477820004,
"max": 0.007265926989012201,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10251584285714288,
"min": 0.10251584285714288,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4352218000000003,
"min": 1.3886848,
"max": 2.6109878,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00026133270142857155,
"min": 0.00026133270142857155,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.003658657820000002,
"min": 0.003658657820000002,
"max": 0.12111768122,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.018223198130726814,
"min": 0.017727622762322426,
"max": 0.5481531023979187,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.2551247775554657,
"min": 0.2551247775554657,
"max": 3.8370718955993652,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 324.68888888888887,
"min": 316.4795918367347,
"max": 990.1176470588235,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 29222.0,
"min": 16832.0,
"max": 32913.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.630848871419827,
"min": -0.9261032789945602,
"max": 1.670409071682529,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 146.77639842778444,
"min": -28.709201648831367,
"max": 162.98439818620682,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.630848871419827,
"min": -0.9261032789945602,
"max": 1.670409071682529,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 146.77639842778444,
"min": -28.709201648831367,
"max": 162.98439818620682,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.06084902421215601,
"min": 0.06084902421215601,
"max": 11.072656408828848,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 5.476412179094041,
"min": 5.228471732989419,
"max": 188.2351589500904,
"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": "1677681614",
"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": "1.22.4",
"end_time_seconds": "1677683914"
},
"total": 2299.832434713,
"count": 1,
"self": 0.4460221419994923,
"children": {
"run_training.setup": {
"total": 0.1219594159999815,
"count": 1,
"self": 0.1219594159999815
},
"TrainerController.start_learning": {
"total": 2299.2644531550004,
"count": 1,
"self": 1.4590554809219611,
"children": {
"TrainerController._reset_env": {
"total": 9.422801222999965,
"count": 1,
"self": 9.422801222999965
},
"TrainerController.advance": {
"total": 2288.298704973078,
"count": 63829,
"self": 1.6340568240566427,
"children": {
"env_step": {
"total": 1547.4688519320125,
"count": 63829,
"self": 1426.2867363680414,
"children": {
"SubprocessEnvManager._take_step": {
"total": 120.28500627400047,
"count": 63829,
"self": 5.034940524973763,
"children": {
"TorchPolicy.evaluate": {
"total": 115.2500657490267,
"count": 62553,
"self": 38.84663123804319,
"children": {
"TorchPolicy.sample_actions": {
"total": 76.40343451098352,
"count": 62553,
"self": 76.40343451098352
}
}
}
}
},
"workers": {
"total": 0.8971092899706719,
"count": 63829,
"self": 0.0,
"children": {
"worker_root": {
"total": 2293.8842412189892,
"count": 63829,
"is_parallel": true,
"self": 991.2021209930072,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.006615324000108558,
"count": 1,
"is_parallel": true,
"self": 0.0038732420000542334,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0027420820000543245,
"count": 8,
"is_parallel": true,
"self": 0.0027420820000543245
}
}
},
"UnityEnvironment.step": {
"total": 0.04790823200005434,
"count": 1,
"is_parallel": true,
"self": 0.000720935999993344,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0005159899999398476,
"count": 1,
"is_parallel": true,
"self": 0.0005159899999398476
},
"communicator.exchange": {
"total": 0.0448776870000529,
"count": 1,
"is_parallel": true,
"self": 0.0448776870000529
},
"steps_from_proto": {
"total": 0.0017936190000682473,
"count": 1,
"is_parallel": true,
"self": 0.0004823159999887139,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0013113030000795334,
"count": 8,
"is_parallel": true,
"self": 0.0013113030000795334
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1302.682120225982,
"count": 63828,
"is_parallel": true,
"self": 32.598678934098416,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 24.817257044948178,
"count": 63828,
"is_parallel": true,
"self": 24.817257044948178
},
"communicator.exchange": {
"total": 1146.2865916779826,
"count": 63828,
"is_parallel": true,
"self": 1146.2865916779826
},
"steps_from_proto": {
"total": 98.97959256895297,
"count": 63828,
"is_parallel": true,
"self": 24.01239839886898,
"children": {
"_process_rank_one_or_two_observation": {
"total": 74.96719417008399,
"count": 510624,
"is_parallel": true,
"self": 74.96719417008399
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 739.1957962170089,
"count": 63829,
"self": 2.8432321060214463,
"children": {
"process_trajectory": {
"total": 170.70768558698535,
"count": 63829,
"self": 170.50410127598514,
"children": {
"RLTrainer._checkpoint": {
"total": 0.20358431100021335,
"count": 2,
"self": 0.20358431100021335
}
}
},
"_update_policy": {
"total": 565.6448785240021,
"count": 457,
"self": 219.40883443200096,
"children": {
"TorchPPOOptimizer.update": {
"total": 346.2360440920012,
"count": 22806,
"self": 346.2360440920012
}
}
}
}
}
}
},
"trainer_threads": {
"total": 9.380000847158954e-07,
"count": 1,
"self": 9.380000847158954e-07
},
"TrainerController._save_models": {
"total": 0.083890540000084,
"count": 1,
"self": 0.0014299360004770278,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08246060399960697,
"count": 1,
"self": 0.08246060399960697
}
}
}
}
}
}
}