ppo-Pyramids / run_logs /timers.json
marimurta's picture
First training of Pyramids
1efbf9c
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.18440984189510345,
"min": 0.17306703329086304,
"max": 0.20108184218406677,
"count": 16
},
"Pyramids.Policy.Entropy.sum": {
"value": 5552.94921875,
"min": 5200.318359375,
"max": 6097.15087890625,
"count": 16
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 242.099173553719,
"min": 229.2741935483871,
"max": 312.4479166666667,
"count": 16
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 29294.0,
"min": 27624.0,
"max": 31868.0,
"count": 16
},
"Pyramids.Step.mean": {
"value": 3479988.0,
"min": 3029872.0,
"max": 3479988.0,
"count": 16
},
"Pyramids.Step.sum": {
"value": 3479988.0,
"min": 3029872.0,
"max": 3479988.0,
"count": 16
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.7435370683670044,
"min": 0.6252047419548035,
"max": 0.8184525966644287,
"count": 16
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 214.138671875,
"min": 176.3077392578125,
"max": 241.44351196289062,
"count": 16
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.013588807545602322,
"min": -0.04507795348763466,
"max": 0.020077427849173546,
"count": 16
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 3.913576602935791,
"min": -13.297996520996094,
"max": 5.782299041748047,
"count": 16
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.759008186029606,
"min": 1.6129184275576212,
"max": 1.768556437785587,
"count": 16
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 214.59899869561195,
"min": 156.0037980079651,
"max": 219.3009982854128,
"count": 16
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.759008186029606,
"min": 1.6129184275576212,
"max": 1.768556437785587,
"count": 16
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 214.59899869561195,
"min": 156.0037980079651,
"max": 219.3009982854128,
"count": 16
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.016038258220894514,
"min": 0.015558057126327185,
"max": 0.020909874858868232,
"count": 16
},
"Pyramids.Policy.RndReward.sum": {
"value": 1.9566675029491307,
"min": 1.8669668551592622,
"max": 2.1741075243626256,
"count": 16
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06727717791329182,
"min": 0.06288980578916299,
"max": 0.07255133635629261,
"count": 16
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9418804907860855,
"min": 0.8804572810482819,
"max": 1.0280237206367449,
"count": 16
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.015126111760143,
"min": 0.012989802343645478,
"max": 0.017220528762360725,
"count": 16
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.211765564642002,
"min": 0.1818572328110367,
"max": 0.24108740267305015,
"count": 16
},
"Pyramids.Policy.LearningRate.mean": {
"value": 2.9299296356346904e-06,
"min": 2.9299296356346904e-06,
"max": 4.150687391950612e-05,
"count": 16
},
"Pyramids.Policy.LearningRate.sum": {
"value": 4.101901489888567e-05,
"min": 4.101901489888567e-05,
"max": 0.0005810962348730857,
"count": 16
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10097661020408162,
"min": 0.10097661020408162,
"max": 0.11383559591836735,
"count": 16
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4136725428571426,
"min": 1.4136725428571426,
"max": 1.6436393428571427,
"count": 16
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00010756335938775502,
"min": 0.00010756335938775502,
"max": 0.001392176032244898,
"count": 16
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0015058870314285703,
"min": 0.0015058870314285703,
"max": 0.01949046445142857,
"count": 16
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.006424341816455126,
"min": 0.006173539441078901,
"max": 0.006839474197477102,
"count": 16
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.08994078636169434,
"min": 0.08642955124378204,
"max": 0.09773396700620651,
"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": "1680095987",
"python_version": "3.9.16 (main, Dec 7 2022, 01:11:51) \n[GCC 9.4.0]",
"command_line_arguments": "/usr/local/bin/mlagents-learn ./config/ppo/PyramidsRND.yaml --resume --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids Training --no-graphics",
"mlagents_version": "0.31.0.dev0",
"mlagents_envs_version": "0.31.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "1.11.0+cu102",
"numpy_version": "1.21.2",
"end_time_seconds": "1680097254"
},
"total": 1266.871840533,
"count": 1,
"self": 0.8468130769997515,
"children": {
"run_training.setup": {
"total": 0.11620823500015831,
"count": 1,
"self": 0.11620823500015831
},
"TrainerController.start_learning": {
"total": 1265.9088192210002,
"count": 1,
"self": 0.7447619251579454,
"children": {
"TrainerController._reset_env": {
"total": 6.05137074599952,
"count": 1,
"self": 6.05137074599952
},
"TrainerController.advance": {
"total": 1258.9613851228423,
"count": 32845,
"self": 0.760670804325855,
"children": {
"env_step": {
"total": 942.2436449983506,
"count": 32845,
"self": 887.6275489564814,
"children": {
"SubprocessEnvManager._take_step": {
"total": 54.19217630285311,
"count": 32845,
"self": 2.3561565757117933,
"children": {
"TorchPolicy.evaluate": {
"total": 51.83601972714132,
"count": 31298,
"self": 51.83601972714132
}
}
},
"workers": {
"total": 0.423919739016128,
"count": 32845,
"self": 0.0,
"children": {
"worker_root": {
"total": 1262.9442338660538,
"count": 32845,
"is_parallel": true,
"self": 435.0136383691279,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0018273859986948082,
"count": 1,
"is_parallel": true,
"self": 0.0005549389970838092,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001272447001610999,
"count": 8,
"is_parallel": true,
"self": 0.001272447001610999
}
}
},
"UnityEnvironment.step": {
"total": 0.049511113998960354,
"count": 1,
"is_parallel": true,
"self": 0.000569089999771677,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00045737099935649894,
"count": 1,
"is_parallel": true,
"self": 0.00045737099935649894
},
"communicator.exchange": {
"total": 0.046839746999467025,
"count": 1,
"is_parallel": true,
"self": 0.046839746999467025
},
"steps_from_proto": {
"total": 0.0016449060003651539,
"count": 1,
"is_parallel": true,
"self": 0.0003675540010590339,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.00127735199930612,
"count": 8,
"is_parallel": true,
"self": 0.00127735199930612
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 827.9305954969259,
"count": 32844,
"is_parallel": true,
"self": 15.859794641866756,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 11.418779589004771,
"count": 32844,
"is_parallel": true,
"self": 11.418779589004771
},
"communicator.exchange": {
"total": 753.328672883119,
"count": 32844,
"is_parallel": true,
"self": 753.328672883119
},
"steps_from_proto": {
"total": 47.323348382935364,
"count": 32844,
"is_parallel": true,
"self": 10.22741873457926,
"children": {
"_process_rank_one_or_two_observation": {
"total": 37.095929648356105,
"count": 262752,
"is_parallel": true,
"self": 37.095929648356105
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 315.95706932016583,
"count": 32845,
"self": 1.4513470632082317,
"children": {
"process_trajectory": {
"total": 62.51331383696379,
"count": 32845,
"self": 62.33615425996322,
"children": {
"RLTrainer._checkpoint": {
"total": 0.17715957700056606,
"count": 1,
"self": 0.17715957700056606
}
}
},
"_update_policy": {
"total": 251.9924084199938,
"count": 236,
"self": 160.5101403259814,
"children": {
"TorchPPOOptimizer.update": {
"total": 91.4822680940124,
"count": 11358,
"self": 91.4822680940124
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.2380005500745028e-06,
"count": 1,
"self": 1.2380005500745028e-06
},
"TrainerController._save_models": {
"total": 0.15130018899981224,
"count": 1,
"self": 0.0034947020012623398,
"children": {
"RLTrainer._checkpoint": {
"total": 0.1478054869985499,
"count": 1,
"self": 0.1478054869985499
}
}
}
}
}
}
}