happycoding's picture
First training of Pyramids Training
ea6552a
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.4022275507450104,
"min": 0.3936680257320404,
"max": 1.4437493085861206,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 11989.5986328125,
"min": 11954.91015625,
"max": 43797.578125,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989968.0,
"min": 29952.0,
"max": 989968.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989968.0,
"min": 29952.0,
"max": 989968.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.43875032663345337,
"min": -0.09125498682260513,
"max": 0.5078338384628296,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 119.340087890625,
"min": -22.17496109008789,
"max": 139.14646911621094,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.013853938318789005,
"min": -0.01204951386898756,
"max": 0.30076801776885986,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 3.7682712078094482,
"min": -3.0967249870300293,
"max": 72.184326171875,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.0670176758063228,
"min": 0.0641688617290042,
"max": 0.07304817292295826,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9382474612885192,
"min": 0.4819602038532003,
"max": 1.0690618422808007,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.01622363798820765,
"min": 0.0010372207016646992,
"max": 0.01806765166977476,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.22713093183490712,
"min": 0.010158061075004695,
"max": 0.26421745702870847,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.723211711342856e-06,
"min": 7.723211711342856e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00010812496395879999,
"min": 0.00010812496395879999,
"max": 0.0033748315750561993,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10257437142857143,
"min": 0.10257437142857143,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4360412,
"min": 1.3691136000000002,
"max": 2.4850423,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0002671797057142857,
"min": 0.0002671797057142857,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0037405158799999998,
"min": 0.0037405158799999998,
"max": 0.11251188562000002,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.017083602026104927,
"min": 0.016848038882017136,
"max": 0.4730415642261505,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.23917041718959808,
"min": 0.2358725517988205,
"max": 3.311290979385376,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 391.4054054054054,
"min": 353.4823529411765,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 28964.0,
"min": 15984.0,
"max": 32676.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.4513626453280448,
"min": -1.0000000521540642,
"max": 1.5968071295037156,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 108.85219839960337,
"min": -32.000001668930054,
"max": 134.1317988783121,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.4513626453280448,
"min": -1.0000000521540642,
"max": 1.5968071295037156,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 108.85219839960337,
"min": -32.000001668930054,
"max": 134.1317988783121,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.06810853450869521,
"min": 0.062566783386934,
"max": 8.998705266043544,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 5.10814008815214,
"min": 5.10814008815214,
"max": 143.9792842566967,
"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": "1673978786",
"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": "1673980729"
},
"total": 1942.846731482,
"count": 1,
"self": 0.49618091299976186,
"children": {
"run_training.setup": {
"total": 0.10928003999998737,
"count": 1,
"self": 0.10928003999998737
},
"TrainerController.start_learning": {
"total": 1942.2412705290003,
"count": 1,
"self": 1.097506983022413,
"children": {
"TrainerController._reset_env": {
"total": 9.758915438999992,
"count": 1,
"self": 9.758915438999992
},
"TrainerController.advance": {
"total": 1931.2952270099784,
"count": 63748,
"self": 1.127215233978859,
"children": {
"env_step": {
"total": 1285.933910555007,
"count": 63748,
"self": 1189.6072198689901,
"children": {
"SubprocessEnvManager._take_step": {
"total": 95.61974447301213,
"count": 63748,
"self": 3.972787641995154,
"children": {
"TorchPolicy.evaluate": {
"total": 91.64695683101698,
"count": 62565,
"self": 31.525925860033084,
"children": {
"TorchPolicy.sample_actions": {
"total": 60.121030970983895,
"count": 62565,
"self": 60.121030970983895
}
}
}
}
},
"workers": {
"total": 0.706946213004727,
"count": 63748,
"self": 0.0,
"children": {
"worker_root": {
"total": 1939.19783875104,
"count": 63748,
"is_parallel": true,
"self": 839.9293963500463,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.010710278999994216,
"count": 1,
"is_parallel": true,
"self": 0.00393215800002622,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.006778120999967996,
"count": 8,
"is_parallel": true,
"self": 0.006778120999967996
}
}
},
"UnityEnvironment.step": {
"total": 0.047256530999987945,
"count": 1,
"is_parallel": true,
"self": 0.0005224720000001071,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0004659789999834629,
"count": 1,
"is_parallel": true,
"self": 0.0004659789999834629
},
"communicator.exchange": {
"total": 0.04463077700000895,
"count": 1,
"is_parallel": true,
"self": 0.04463077700000895
},
"steps_from_proto": {
"total": 0.001637302999995427,
"count": 1,
"is_parallel": true,
"self": 0.00041433099997334466,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0012229720000220823,
"count": 8,
"is_parallel": true,
"self": 0.0012229720000220823
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1099.2684424009938,
"count": 63747,
"is_parallel": true,
"self": 27.342294746978496,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 21.813193095005175,
"count": 63747,
"is_parallel": true,
"self": 21.813193095005175
},
"communicator.exchange": {
"total": 961.275220771993,
"count": 63747,
"is_parallel": true,
"self": 961.275220771993
},
"steps_from_proto": {
"total": 88.8377337870171,
"count": 63747,
"is_parallel": true,
"self": 20.345982458104345,
"children": {
"_process_rank_one_or_two_observation": {
"total": 68.49175132891276,
"count": 509976,
"is_parallel": true,
"self": 68.49175132891276
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 644.2341012209924,
"count": 63748,
"self": 2.0717154819753887,
"children": {
"process_trajectory": {
"total": 139.55135413701822,
"count": 63748,
"self": 139.3616066040185,
"children": {
"RLTrainer._checkpoint": {
"total": 0.1897475329997178,
"count": 2,
"self": 0.1897475329997178
}
}
},
"_update_policy": {
"total": 502.6110316019988,
"count": 444,
"self": 194.54644849800172,
"children": {
"TorchPPOOptimizer.update": {
"total": 308.06458310399705,
"count": 22803,
"self": 308.06458310399705
}
}
}
}
}
}
},
"trainer_threads": {
"total": 7.139997251215391e-07,
"count": 1,
"self": 7.139997251215391e-07
},
"TrainerController._save_models": {
"total": 0.08962038299978303,
"count": 1,
"self": 0.001557903999582777,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08806247900020026,
"count": 1,
"self": 0.08806247900020026
}
}
}
}
}
}
}