ppo-Pyramids / run_logs /timers.json
RegisGraptin's picture
First training
5e15e80
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.3682090938091278,
"min": 0.3682090938091278,
"max": 1.4500666856765747,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 11099.294921875,
"min": 11086.451171875,
"max": 43989.22265625,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989894.0,
"min": 29952.0,
"max": 989894.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989894.0,
"min": 29952.0,
"max": 989894.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.19989311695098877,
"min": -0.09076467156410217,
"max": 0.23905643820762634,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 51.172637939453125,
"min": -21.874286651611328,
"max": 60.95939254760742,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.0413900688290596,
"min": 0.002438932191580534,
"max": 0.2935013175010681,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 10.595857620239258,
"min": 0.6219277381896973,
"max": 69.559814453125,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06363395912526328,
"min": 0.06363395912526328,
"max": 0.07142895846717498,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.954509386878949,
"min": 0.5000027092702248,
"max": 1.0553752826429748,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.010151777975109034,
"min": 0.00041157038140970107,
"max": 0.010546946605245466,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.15227666962663552,
"min": 0.004527274195506712,
"max": 0.15227666962663552,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.512237495953333e-06,
"min": 7.512237495953333e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.0001126835624393,
"min": 0.0001126835624393,
"max": 0.003508912430362599,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10250404666666668,
"min": 0.10250404666666668,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.5375607000000002,
"min": 1.3886848,
"max": 2.5696374,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00026015426200000005,
"min": 0.00026015426200000005,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.003902313930000001,
"min": 0.003902313930000001,
"max": 0.11698677626000002,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.01082004327327013,
"min": 0.01082004327327013,
"max": 0.3558875620365143,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.16230064630508423,
"min": 0.15213251113891602,
"max": 2.491212844848633,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 581.92,
"min": 581.92,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 29096.0,
"min": 15984.0,
"max": 33299.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 0.937935971468687,
"min": -1.0000000521540642,
"max": 0.9446038115196503,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 46.89679857343435,
"min": -31.998801663517952,
"max": 49.119398199021816,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 0.937935971468687,
"min": -1.0000000521540642,
"max": 0.9446038115196503,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 46.89679857343435,
"min": -31.998801663517952,
"max": 49.119398199021816,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.06411020807834575,
"min": 0.06411020807834575,
"max": 6.648574652150273,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 3.2055104039172875,
"min": 3.2055104039172875,
"max": 106.37719443440437,
"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": "1673778156",
"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": "1673780028"
},
"total": 1872.5614553860003,
"count": 1,
"self": 0.4243667540001752,
"children": {
"run_training.setup": {
"total": 0.10473625300005551,
"count": 1,
"self": 0.10473625300005551
},
"TrainerController.start_learning": {
"total": 1872.032352379,
"count": 1,
"self": 1.091142634040807,
"children": {
"TrainerController._reset_env": {
"total": 6.316977988999952,
"count": 1,
"self": 6.316977988999952
},
"TrainerController.advance": {
"total": 1864.5342170719593,
"count": 63320,
"self": 1.157561034929813,
"children": {
"env_step": {
"total": 1227.2059674540124,
"count": 63320,
"self": 1129.8963595100254,
"children": {
"SubprocessEnvManager._take_step": {
"total": 96.6067285479337,
"count": 63320,
"self": 4.0151677429839765,
"children": {
"TorchPolicy.evaluate": {
"total": 92.59156080494972,
"count": 62566,
"self": 31.524634768910573,
"children": {
"TorchPolicy.sample_actions": {
"total": 61.06692603603915,
"count": 62566,
"self": 61.06692603603915
}
}
}
}
},
"workers": {
"total": 0.7028793960532767,
"count": 63320,
"self": 0.0,
"children": {
"worker_root": {
"total": 1868.240944596031,
"count": 63320,
"is_parallel": true,
"self": 830.0069625280689,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0017041639998751634,
"count": 1,
"is_parallel": true,
"self": 0.0006504069997390616,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0010537570001361019,
"count": 8,
"is_parallel": true,
"self": 0.0010537570001361019
}
}
},
"UnityEnvironment.step": {
"total": 0.043419805999974415,
"count": 1,
"is_parallel": true,
"self": 0.0005145090001406061,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00042572899997139757,
"count": 1,
"is_parallel": true,
"self": 0.00042572899997139757
},
"communicator.exchange": {
"total": 0.0408647489998657,
"count": 1,
"is_parallel": true,
"self": 0.0408647489998657
},
"steps_from_proto": {
"total": 0.0016148189999967144,
"count": 1,
"is_parallel": true,
"self": 0.00039532899972982705,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0012194900002668874,
"count": 8,
"is_parallel": true,
"self": 0.0012194900002668874
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1038.2339820679622,
"count": 63319,
"is_parallel": true,
"self": 27.027536439902633,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 21.81387560506323,
"count": 63319,
"is_parallel": true,
"self": 21.81387560506323
},
"communicator.exchange": {
"total": 892.5962323390195,
"count": 63319,
"is_parallel": true,
"self": 892.5962323390195
},
"steps_from_proto": {
"total": 96.79633768397684,
"count": 63319,
"is_parallel": true,
"self": 20.98916796788035,
"children": {
"_process_rank_one_or_two_observation": {
"total": 75.8071697160965,
"count": 506552,
"is_parallel": true,
"self": 75.8071697160965
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 636.1706885830172,
"count": 63320,
"self": 2.095910631022889,
"children": {
"process_trajectory": {
"total": 136.52222994699696,
"count": 63320,
"self": 136.32273886299754,
"children": {
"RLTrainer._checkpoint": {
"total": 0.19949108399941906,
"count": 2,
"self": 0.19949108399941906
}
}
},
"_update_policy": {
"total": 497.55254800499733,
"count": 451,
"self": 192.05160618096352,
"children": {
"TorchPPOOptimizer.update": {
"total": 305.5009418240338,
"count": 22764,
"self": 305.5009418240338
}
}
}
}
}
}
},
"trainer_threads": {
"total": 9.350001164420974e-07,
"count": 1,
"self": 9.350001164420974e-07
},
"TrainerController._save_models": {
"total": 0.09001374899980874,
"count": 1,
"self": 0.001648159999604104,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08836558900020464,
"count": 1,
"self": 0.08836558900020464
}
}
}
}
}
}
}