ppo-PyramidsRND / run_logs /timers.json
michal512's picture
First Pyramid
08e5906
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.5277588963508606,
"min": 0.5177602171897888,
"max": 1.4129016399383545,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 15765.212890625,
"min": 15524.521484375,
"max": 42861.78515625,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989990.0,
"min": 29952.0,
"max": 989990.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989990.0,
"min": 29952.0,
"max": 989990.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.4553857147693634,
"min": -0.10661628842353821,
"max": 0.46414968371391296,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 121.58798217773438,
"min": -25.69452476501465,
"max": 122.99966430664062,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.02170846238732338,
"min": -0.08050467818975449,
"max": 0.19064931571483612,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 5.796159267425537,
"min": -21.333740234375,
"max": 45.94648361206055,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06724398330681854,
"min": 0.0651322675206374,
"max": 0.07427928427771095,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 1.008659749602278,
"min": 0.5171115858212378,
"max": 1.1141892641656643,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.014928267443449134,
"min": 8.691777735898267e-05,
"max": 0.01614573707503915,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.223924011651737,
"min": 0.0011299311056667746,
"max": 0.22604031905054808,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.406477531206668e-06,
"min": 7.406477531206668e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00011109716296810002,
"min": 0.00011109716296810002,
"max": 0.0032555372148209997,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10246879333333334,
"min": 0.10246879333333334,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.5370319000000001,
"min": 1.3886848,
"max": 2.4375519,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0002566324540000001,
"min": 0.0002566324540000001,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.003849486810000001,
"min": 0.003849486810000001,
"max": 0.1085393821,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.010050839744508266,
"min": 0.00944732129573822,
"max": 0.3706575036048889,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.15076260268688202,
"min": 0.13226249814033508,
"max": 2.594602584838867,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 400.4189189189189,
"min": 400.4189189189189,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 29631.0,
"min": 15984.0,
"max": 32991.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.4700426406661669,
"min": -1.0000000521540642,
"max": 1.5242901218818947,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 110.25319804996252,
"min": -32.000001668930054,
"max": 110.25319804996252,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.4700426406661669,
"min": -1.0000000521540642,
"max": 1.5242901218818947,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 110.25319804996252,
"min": -32.000001668930054,
"max": 110.25319804996252,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.04133921385121842,
"min": 0.04133921385121842,
"max": 7.576801156625152,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 3.1004410388413817,
"min": 2.987983369093854,
"max": 121.22881850600243,
"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": "1676564216",
"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": "1.21.6",
"end_time_seconds": "1676567743"
},
"total": 3526.673220793,
"count": 1,
"self": 1.4717146520006281,
"children": {
"run_training.setup": {
"total": 0.1407273229999646,
"count": 1,
"self": 0.1407273229999646
},
"TrainerController.start_learning": {
"total": 3525.0607788179996,
"count": 1,
"self": 2.1396914010242654,
"children": {
"TrainerController._reset_env": {
"total": 6.5255688910000345,
"count": 1,
"self": 6.5255688910000345
},
"TrainerController.advance": {
"total": 3516.2254349539753,
"count": 63553,
"self": 2.4030813110266536,
"children": {
"env_step": {
"total": 2235.0674829779746,
"count": 63553,
"self": 2089.2447677058926,
"children": {
"SubprocessEnvManager._take_step": {
"total": 144.35277112600602,
"count": 63553,
"self": 6.964825576972601,
"children": {
"TorchPolicy.evaluate": {
"total": 137.38794554903342,
"count": 62557,
"self": 29.239629377015376,
"children": {
"TorchPolicy.sample_actions": {
"total": 108.14831617201804,
"count": 62557,
"self": 108.14831617201804
}
}
}
}
},
"workers": {
"total": 1.4699441460761022,
"count": 63553,
"self": 0.0,
"children": {
"worker_root": {
"total": 3517.6248623950305,
"count": 63553,
"is_parallel": true,
"self": 1607.1051898140704,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.006804117000001497,
"count": 1,
"is_parallel": true,
"self": 0.003971591999970769,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.002832525000030728,
"count": 8,
"is_parallel": true,
"self": 0.002832525000030728
}
}
},
"UnityEnvironment.step": {
"total": 0.06149703400001272,
"count": 1,
"is_parallel": true,
"self": 0.0007326740000621612,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00033016300000099363,
"count": 1,
"is_parallel": true,
"self": 0.00033016300000099363
},
"communicator.exchange": {
"total": 0.05825534999996762,
"count": 1,
"is_parallel": true,
"self": 0.05825534999996762
},
"steps_from_proto": {
"total": 0.0021788469999819426,
"count": 1,
"is_parallel": true,
"self": 0.0004919049999898562,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0016869419999920865,
"count": 8,
"is_parallel": true,
"self": 0.0016869419999920865
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1910.51967258096,
"count": 63552,
"is_parallel": true,
"self": 47.65770180496429,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 26.752699171047084,
"count": 63552,
"is_parallel": true,
"self": 26.752699171047084
},
"communicator.exchange": {
"total": 1699.9535816159764,
"count": 63552,
"is_parallel": true,
"self": 1699.9535816159764
},
"steps_from_proto": {
"total": 136.15568998897237,
"count": 63552,
"is_parallel": true,
"self": 33.485671846996865,
"children": {
"_process_rank_one_or_two_observation": {
"total": 102.6700181419755,
"count": 508416,
"is_parallel": true,
"self": 102.6700181419755
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 1278.7548706649743,
"count": 63553,
"self": 4.441461429927358,
"children": {
"process_trajectory": {
"total": 226.63636701405142,
"count": 63553,
"self": 226.22112562305085,
"children": {
"RLTrainer._checkpoint": {
"total": 0.41524139100056345,
"count": 2,
"self": 0.41524139100056345
}
}
},
"_update_policy": {
"total": 1047.6770422209956,
"count": 444,
"self": 280.4397256300034,
"children": {
"TorchPPOOptimizer.update": {
"total": 767.2373165909922,
"count": 22794,
"self": 767.2373165909922
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.497000084782485e-06,
"count": 1,
"self": 1.497000084782485e-06
},
"TrainerController._save_models": {
"total": 0.1700820749997547,
"count": 1,
"self": 0.005205556999953842,
"children": {
"RLTrainer._checkpoint": {
"total": 0.16487651799980085,
"count": 1,
"self": 0.16487651799980085
}
}
}
}
}
}
}