ppo-Pyramids / run_logs /timers.json
cyeet's picture
First training
15a4b67
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.38535264134407043,
"min": 0.38535264134407043,
"max": 1.4340118169784546,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 11622.2353515625,
"min": 11515.4736328125,
"max": 43502.18359375,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989971.0,
"min": 29901.0,
"max": 989971.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989971.0,
"min": 29901.0,
"max": 989971.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.643875777721405,
"min": -0.10255663841962814,
"max": 0.643875777721405,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 184.79234313964844,
"min": -24.818706512451172,
"max": 184.79234313964844,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.010809449478983879,
"min": -0.005641380324959755,
"max": 0.6432042717933655,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 3.1023120880126953,
"min": -1.5457382202148438,
"max": 152.4394073486328,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06856560069302628,
"min": 0.06486136321870596,
"max": 0.07316504085871962,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.959918409702368,
"min": 0.4832980583188386,
"max": 1.0612731124685308,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.015947685946316813,
"min": 0.0005039816814433899,
"max": 0.026813511493280946,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.22326760324843536,
"min": 0.006551761858764068,
"max": 0.24162014279863797,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.564854621271427e-06,
"min": 7.564854621271427e-06,
"max": 0.0002952351015883,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00010590796469779998,
"min": 0.00010590796469779998,
"max": 0.0033802268732578008,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10252158571428571,
"min": 0.10252158571428571,
"max": 0.1984117,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4353022,
"min": 1.3888819,
"max": 2.5267422,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00026190641285714285,
"min": 0.00026190641285714285,
"max": 0.00984132883,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.00366668978,
"min": 0.00366668978,
"max": 0.11270154578000001,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.01548650860786438,
"min": 0.01548650860786438,
"max": 0.8159781694412231,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.21681112051010132,
"min": 0.21681112051010132,
"max": 5.711847305297852,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 287.3047619047619,
"min": 287.3047619047619,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 30167.0,
"min": 16700.0,
"max": 32432.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.6745885523302215,
"min": -0.999987552408129,
"max": 1.6745885523302215,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 175.83179799467325,
"min": -31.999601677060127,
"max": 175.83179799467325,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.6745885523302215,
"min": -0.999987552408129,
"max": 1.6745885523302215,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 175.83179799467325,
"min": -31.999601677060127,
"max": 175.83179799467325,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.04611302747840195,
"min": 0.04611302747840195,
"max": 17.58480724341729,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 4.841867885232205,
"min": 4.660183200932806,
"max": 298.94172313809395,
"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": "1673497079",
"python_version": "3.8.16 (default, Dec 7 2022, 01:12:13) \n[GCC 7.5.0]",
"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": "1673499111"
},
"total": 2032.4949281039999,
"count": 1,
"self": 0.384770509999953,
"children": {
"run_training.setup": {
"total": 0.10326109800007544,
"count": 1,
"self": 0.10326109800007544
},
"TrainerController.start_learning": {
"total": 2032.0068964959999,
"count": 1,
"self": 1.2466524079970895,
"children": {
"TrainerController._reset_env": {
"total": 6.2558559799999784,
"count": 1,
"self": 6.2558559799999784
},
"TrainerController.advance": {
"total": 2024.4139960590023,
"count": 63947,
"self": 1.2847530909477882,
"children": {
"env_step": {
"total": 1366.9868217920723,
"count": 63947,
"self": 1264.833158251176,
"children": {
"SubprocessEnvManager._take_step": {
"total": 101.37037233291721,
"count": 63947,
"self": 4.2471907519106935,
"children": {
"TorchPolicy.evaluate": {
"total": 97.12318158100652,
"count": 62560,
"self": 32.473596955978564,
"children": {
"TorchPolicy.sample_actions": {
"total": 64.64958462502796,
"count": 62560,
"self": 64.64958462502796
}
}
}
}
},
"workers": {
"total": 0.7832912079791186,
"count": 63947,
"self": 0.0,
"children": {
"worker_root": {
"total": 2027.7310040110192,
"count": 63947,
"is_parallel": true,
"self": 860.8068762800001,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.00181788199984112,
"count": 1,
"is_parallel": true,
"self": 0.0006273579999742651,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011905239998668549,
"count": 8,
"is_parallel": true,
"self": 0.0011905239998668549
}
}
},
"UnityEnvironment.step": {
"total": 0.04761529399979736,
"count": 1,
"is_parallel": true,
"self": 0.0004845369999202376,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00044205399990460137,
"count": 1,
"is_parallel": true,
"self": 0.00044205399990460137
},
"communicator.exchange": {
"total": 0.04513455899996188,
"count": 1,
"is_parallel": true,
"self": 0.04513455899996188
},
"steps_from_proto": {
"total": 0.0015541440000106377,
"count": 1,
"is_parallel": true,
"self": 0.00041471899930911604,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011394250007015216,
"count": 8,
"is_parallel": true,
"self": 0.0011394250007015216
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1166.924127731019,
"count": 63946,
"is_parallel": true,
"self": 27.618510007978557,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 22.577149495026788,
"count": 63946,
"is_parallel": true,
"self": 22.577149495026788
},
"communicator.exchange": {
"total": 1017.1851497929922,
"count": 63946,
"is_parallel": true,
"self": 1017.1851497929922
},
"steps_from_proto": {
"total": 99.54331843502155,
"count": 63946,
"is_parallel": true,
"self": 21.79434892397103,
"children": {
"_process_rank_one_or_two_observation": {
"total": 77.74896951105052,
"count": 511568,
"is_parallel": true,
"self": 77.74896951105052
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 656.1424211759822,
"count": 63947,
"self": 2.426197324955865,
"children": {
"process_trajectory": {
"total": 142.53838653202888,
"count": 63947,
"self": 142.33866893302934,
"children": {
"RLTrainer._checkpoint": {
"total": 0.19971759899954122,
"count": 2,
"self": 0.19971759899954122
}
}
},
"_update_policy": {
"total": 511.1778373189975,
"count": 452,
"self": 199.35278211000355,
"children": {
"TorchPPOOptimizer.update": {
"total": 311.82505520899394,
"count": 22779,
"self": 311.82505520899394
}
}
}
}
}
}
},
"trainer_threads": {
"total": 9.300001693191007e-07,
"count": 1,
"self": 9.300001693191007e-07
},
"TrainerController._save_models": {
"total": 0.09039111900028729,
"count": 1,
"self": 0.0014103810003689432,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08898073799991835,
"count": 1,
"self": 0.08898073799991835
}
}
}
}
}
}
}