ppo-Pyramids / run_logs /timers.json
albertqueralto's picture
First Push
98f5890
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.693298876285553,
"min": 0.693298876285553,
"max": 1.376702070236206,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 20743.501953125,
"min": 20743.501953125,
"max": 41763.6328125,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989950.0,
"min": 29876.0,
"max": 989950.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989950.0,
"min": 29876.0,
"max": 989950.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.32271119952201843,
"min": -0.10292070358991623,
"max": 0.32862532138824463,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 83.25949096679688,
"min": -24.803890228271484,
"max": 85.77120971679688,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.020750124007463455,
"min": 0.012273435480892658,
"max": 0.3880854845046997,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 5.353531837463379,
"min": 3.103821277618408,
"max": 91.97625732421875,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06626238100791901,
"min": 0.062330088698819106,
"max": 0.07280406456284634,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.927673334110866,
"min": 0.5721240846597363,
"max": 1.0498490698519163,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.011780233855038218,
"min": 0.00062098568645058,
"max": 0.018500328734870317,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.16492327397053505,
"min": 0.00869379961030812,
"max": 0.19927606230667866,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.243254728471429e-06,
"min": 7.243254728471429e-06,
"max": 0.00029484855171715,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.0001014055661986,
"min": 0.0001014055661986,
"max": 0.0037572196475934992,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10241438571428572,
"min": 0.10241438571428572,
"max": 0.19828284999999998,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4338014000000001,
"min": 1.4338014000000001,
"max": 2.6524065,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00025119713285714286,
"min": 0.00025119713285714286,
"max": 0.009828456715,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0035167598600000004,
"min": 0.0035167598600000004,
"max": 0.12525540935,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.0083738062530756,
"min": 0.0083738062530756,
"max": 0.6670273542404175,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.1172332912683487,
"min": 0.1172332912683487,
"max": 5.33621883392334,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 590.3461538461538,
"min": 509.5,
"max": 977.8181818181819,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 30698.0,
"min": 16419.0,
"max": 32493.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.1018461175262928,
"min": -0.9181576280882864,
"max": 1.237540318254839,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 57.295998111367226,
"min": -30.299201726913452,
"max": 70.53979814052582,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.1018461175262928,
"min": -0.9181576280882864,
"max": 1.237540318254839,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 57.295998111367226,
"min": -30.299201726913452,
"max": 70.53979814052582,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.05208041314640118,
"min": 0.046122753381496295,
"max": 13.765640188665952,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 2.7081814836128615,
"min": 2.675119696126785,
"max": 234.01588320732117,
"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": "1677693692",
"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": "1677695816"
},
"total": 2124.564268623,
"count": 1,
"self": 0.5414670709997154,
"children": {
"run_training.setup": {
"total": 0.19648140900000044,
"count": 1,
"self": 0.19648140900000044
},
"TrainerController.start_learning": {
"total": 2123.826320143,
"count": 1,
"self": 1.3151167769910899,
"children": {
"TrainerController._reset_env": {
"total": 8.467082976000029,
"count": 1,
"self": 8.467082976000029
},
"TrainerController.advance": {
"total": 2113.961875099009,
"count": 63395,
"self": 1.4111754480395575,
"children": {
"env_step": {
"total": 1386.9253384029898,
"count": 63395,
"self": 1273.8235399310167,
"children": {
"SubprocessEnvManager._take_step": {
"total": 112.30239431898957,
"count": 63395,
"self": 4.62383089996456,
"children": {
"TorchPolicy.evaluate": {
"total": 107.67856341902501,
"count": 62554,
"self": 36.66447431103785,
"children": {
"TorchPolicy.sample_actions": {
"total": 71.01408910798716,
"count": 62554,
"self": 71.01408910798716
}
}
}
}
},
"workers": {
"total": 0.799404152983584,
"count": 63395,
"self": 0.0,
"children": {
"worker_root": {
"total": 2119.5314955540243,
"count": 63395,
"is_parallel": true,
"self": 957.9786714930397,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0046775010000033035,
"count": 1,
"is_parallel": true,
"self": 0.003553383999928883,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011241170000744205,
"count": 8,
"is_parallel": true,
"self": 0.0011241170000744205
}
}
},
"UnityEnvironment.step": {
"total": 0.04644266700000799,
"count": 1,
"is_parallel": true,
"self": 0.000705508999999438,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0005316349999588965,
"count": 1,
"is_parallel": true,
"self": 0.0005316349999588965
},
"communicator.exchange": {
"total": 0.04354102400003512,
"count": 1,
"is_parallel": true,
"self": 0.04354102400003512
},
"steps_from_proto": {
"total": 0.0016644990000145299,
"count": 1,
"is_parallel": true,
"self": 0.000416800999971656,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001247698000042874,
"count": 8,
"is_parallel": true,
"self": 0.001247698000042874
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1161.5528240609847,
"count": 63394,
"is_parallel": true,
"self": 31.101328546908235,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 22.59144627000188,
"count": 63394,
"is_parallel": true,
"self": 22.59144627000188
},
"communicator.exchange": {
"total": 1018.9309719680291,
"count": 63394,
"is_parallel": true,
"self": 1018.9309719680291
},
"steps_from_proto": {
"total": 88.92907727604546,
"count": 63394,
"is_parallel": true,
"self": 20.798063996073438,
"children": {
"_process_rank_one_or_two_observation": {
"total": 68.13101327997202,
"count": 507152,
"is_parallel": true,
"self": 68.13101327997202
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 725.6253612479795,
"count": 63395,
"self": 2.546306871945717,
"children": {
"process_trajectory": {
"total": 162.1085195980337,
"count": 63395,
"self": 161.92147594003347,
"children": {
"RLTrainer._checkpoint": {
"total": 0.1870436580002206,
"count": 2,
"self": 0.1870436580002206
}
}
},
"_update_policy": {
"total": 560.9705347780001,
"count": 454,
"self": 213.49560886302828,
"children": {
"TorchPPOOptimizer.update": {
"total": 347.47492591497183,
"count": 22734,
"self": 347.47492591497183
}
}
}
}
}
}
},
"trainer_threads": {
"total": 8.049996722547803e-07,
"count": 1,
"self": 8.049996722547803e-07
},
"TrainerController._save_models": {
"total": 0.08224448600003598,
"count": 1,
"self": 0.0014033510001354443,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08084113499990053,
"count": 1,
"self": 0.08084113499990053
}
}
}
}
}
}
}