ppo-Pyramids / run_logs /timers.json
crossroderick's picture
First training of Pyramids
2b2759a verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.5221751928329468,
"min": 0.5221751928329468,
"max": 1.454694151878357,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 15606.771484375,
"min": 15606.771484375,
"max": 44129.6015625,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989877.0,
"min": 29952.0,
"max": 989877.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989877.0,
"min": 29952.0,
"max": 989877.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.43396618962287903,
"min": -0.22855108976364136,
"max": 0.43814659118652344,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 115.43500518798828,
"min": -54.16660690307617,
"max": 118.29957580566406,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": -0.029414575546979904,
"min": -0.05852945148944855,
"max": 0.2471713274717331,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": -7.824276924133301,
"min": -15.685893058776855,
"max": 58.57960510253906,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.07159054563293547,
"min": 0.06598144438785745,
"max": 0.07391394989763128,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 1.0022676388610967,
"min": 0.4984373760623082,
"max": 1.083893627106947,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.015346981370578657,
"min": 0.00045631518334628834,
"max": 0.01699009824023094,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.2148577391881012,
"min": 0.0045631518334628834,
"max": 0.2406875241116116,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.729768852014285e-06,
"min": 7.729768852014285e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00010821676392819998,
"min": 0.00010821676392819998,
"max": 0.0035076956307681997,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10257655714285718,
"min": 0.10257655714285718,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4360718000000006,
"min": 1.3886848,
"max": 2.5692318000000007,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00026739805857142854,
"min": 0.00026739805857142854,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0037435728199999995,
"min": 0.0037435728199999995,
"max": 0.11694625682000002,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.00871711503714323,
"min": 0.0081988126039505,
"max": 0.3385855555534363,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.12203960865736008,
"min": 0.114783376455307,
"max": 2.370098829269409,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 429.1044776119403,
"min": 395.13698630136986,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 28750.0,
"min": 15984.0,
"max": 33672.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.541029834814036,
"min": -1.0000000521540642,
"max": 1.541029834814036,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 103.24899893254042,
"min": -31.998801663517952,
"max": 109.68039886653423,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.541029834814036,
"min": -1.0000000521540642,
"max": 1.541029834814036,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 103.24899893254042,
"min": -31.998801663517952,
"max": 109.68039886653423,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.038530179310599585,
"min": 0.03696660857571664,
"max": 6.70474449545145,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 2.581522013810172,
"min": 2.5363514529162785,
"max": 107.2759119272232,
"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": "1709679820",
"python_version": "3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]",
"command_line_arguments": "/home/rodrigopc/.local/bin/mlagents-learn PyramidsRND_modified.yaml --env=./training-envs-executables/linux/Pyramids/Pyramids.app --run-id=Pyramids1 --no-graphics",
"mlagents_version": "1.1.0.dev0",
"mlagents_envs_version": "1.1.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.2.1+cu121",
"numpy_version": "1.23.5",
"end_time_seconds": "1709681525"
},
"total": 1704.454596974996,
"count": 1,
"self": 0.42225266899185954,
"children": {
"run_training.setup": {
"total": 0.05185863099904964,
"count": 1,
"self": 0.05185863099904964
},
"TrainerController.start_learning": {
"total": 1703.9804856750052,
"count": 1,
"self": 1.168904307167395,
"children": {
"TrainerController._reset_env": {
"total": 3.5841346029992565,
"count": 1,
"self": 3.5841346029992565
},
"TrainerController.advance": {
"total": 1699.1427650148398,
"count": 63481,
"self": 1.2076140538556501,
"children": {
"env_step": {
"total": 1173.4178847450821,
"count": 63481,
"self": 1069.1849527462473,
"children": {
"SubprocessEnvManager._take_step": {
"total": 103.41836099699867,
"count": 63481,
"self": 4.307201075687772,
"children": {
"TorchPolicy.evaluate": {
"total": 99.1111599213109,
"count": 62567,
"self": 99.1111599213109
}
}
},
"workers": {
"total": 0.8145710018361569,
"count": 63481,
"self": 0.0,
"children": {
"worker_root": {
"total": 1701.7859792481977,
"count": 63481,
"is_parallel": true,
"self": 726.6117202753157,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.002736848997301422,
"count": 1,
"is_parallel": true,
"self": 0.0007465309972758405,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0019903180000255816,
"count": 8,
"is_parallel": true,
"self": 0.0019903180000255816
}
}
},
"UnityEnvironment.step": {
"total": 0.04309753300185548,
"count": 1,
"is_parallel": true,
"self": 0.0003881740049109794,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0003291589964646846,
"count": 1,
"is_parallel": true,
"self": 0.0003291589964646846
},
"communicator.exchange": {
"total": 0.041251364003983326,
"count": 1,
"is_parallel": true,
"self": 0.041251364003983326
},
"steps_from_proto": {
"total": 0.0011288359964964911,
"count": 1,
"is_parallel": true,
"self": 0.00023969398898771033,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0008891420075087808,
"count": 8,
"is_parallel": true,
"self": 0.0008891420075087808
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 975.174258972882,
"count": 63480,
"is_parallel": true,
"self": 31.38852427239908,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 24.111143964502844,
"count": 63480,
"is_parallel": true,
"self": 24.111143964502844
},
"communicator.exchange": {
"total": 840.062954167086,
"count": 63480,
"is_parallel": true,
"self": 840.062954167086
},
"steps_from_proto": {
"total": 79.61163656889403,
"count": 63480,
"is_parallel": true,
"self": 16.43438122081716,
"children": {
"_process_rank_one_or_two_observation": {
"total": 63.17725534807687,
"count": 507840,
"is_parallel": true,
"self": 63.17725534807687
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 524.517266215902,
"count": 63481,
"self": 2.4202581968856975,
"children": {
"process_trajectory": {
"total": 79.72593308304931,
"count": 63481,
"self": 79.52337400104443,
"children": {
"RLTrainer._checkpoint": {
"total": 0.20255908200488193,
"count": 2,
"self": 0.20255908200488193
}
}
},
"_update_policy": {
"total": 442.37107493596704,
"count": 445,
"self": 227.37120656114712,
"children": {
"TorchPPOOptimizer.update": {
"total": 214.99986837481993,
"count": 22827,
"self": 214.99986837481993
}
}
}
}
}
}
},
"trainer_threads": {
"total": 2.373999450355768e-06,
"count": 1,
"self": 2.373999450355768e-06
},
"TrainerController._save_models": {
"total": 0.0846793759992579,
"count": 1,
"self": 0.002215993001300376,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08246338299795752,
"count": 1,
"self": 0.08246338299795752
}
}
}
}
}
}
}