ppo-Pyramids / run_logs /timers.json
magnustragardh's picture
9 million training steps
15a50f8
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.12155546993017197,
"min": 0.10337793081998825,
"max": 0.166367307305336,
"count": 100
},
"Pyramids.Policy.Entropy.sum": {
"value": 3617.49072265625,
"min": 3098.02978515625,
"max": 5028.28564453125,
"count": 100
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 203.24137931034483,
"min": 178.5988023952096,
"max": 276.77064220183485,
"count": 100
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 29470.0,
"min": 26913.0,
"max": 32611.0,
"count": 100
},
"Pyramids.Step.mean": {
"value": 8999992.0,
"min": 6029924.0,
"max": 8999992.0,
"count": 100
},
"Pyramids.Step.sum": {
"value": 8999992.0,
"min": 6029924.0,
"max": 8999992.0,
"count": 100
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.8813865184783936,
"min": 0.7026673555374146,
"max": 0.9559917449951172,
"count": 100
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 274.1112060546875,
"min": 203.07086181640625,
"max": 306.87335205078125,
"count": 100
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.00841159000992775,
"min": -0.0707196518778801,
"max": 0.032943692058324814,
"count": 100
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 2.616004467010498,
"min": -21.07445526123047,
"max": 9.718389511108398,
"count": 100
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.7829641251728452,
"min": 1.57638714865807,
"max": 1.8214011828521055,
"count": 100
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 258.52979815006256,
"min": 171.82619920372963,
"max": 304.1739975363016,
"count": 100
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.7829641251728452,
"min": 1.57638714865807,
"max": 1.8214011828521055,
"count": 100
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 258.52979815006256,
"min": 171.82619920372963,
"max": 304.1739975363016,
"count": 100
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.009233273781538973,
"min": 0.008271286948001944,
"max": 0.014159980635803331,
"count": 100
},
"Pyramids.Policy.RndReward.sum": {
"value": 1.338824698323151,
"min": 1.228675196536642,
"max": 1.861013156008994,
"count": 100
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06888185016023705,
"min": 0.06377542618380375,
"max": 0.07488835589278356,
"count": 100
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9643459022433187,
"min": 0.8928559665732525,
"max": 1.094879996866974,
"count": 100
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.01753181091533238,
"min": 0.01119533650635276,
"max": 0.01753181091533238,
"count": 100
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.24544535281465335,
"min": 0.16624239571655042,
"max": 0.24544535281465335,
"count": 100
},
"Pyramids.Policy.LearningRate.mean": {
"value": 4.860760284841276e-07,
"min": 4.860760284841276e-07,
"max": 9.947281446146507e-05,
"count": 100
},
"Pyramids.Policy.LearningRate.sum": {
"value": 6.805064398777786e-06,
"min": 6.805064398777786e-06,
"max": 0.0014474719841763442,
"count": 100
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10016199206349205,
"min": 0.10016199206349205,
"max": 0.13315758253968255,
"count": 100
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4022678888888886,
"min": 1.4022678888888886,
"max": 1.9824903222222225,
"count": 100
},
"Pyramids.Policy.Beta.mean": {
"value": 2.6183007142857167e-05,
"min": 2.6183007142857167e-05,
"max": 0.0033224424957142866,
"count": 100
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0003665621000000003,
"min": 0.0003665621000000003,
"max": 0.04835078319000001,
"count": 100
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.00432272395119071,
"min": 0.003942703828215599,
"max": 0.005807704292237759,
"count": 100
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.060518134385347366,
"min": 0.057913638651371,
"max": 0.08529843389987946,
"count": 100
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 100
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 100
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1688313673",
"python_version": "3.10.12 (main, Jun 9 2023, 20:10:04) [GCC 13.1.1 20230522 [revision dd36656ada05731c069ecd5b1878380294fb1f3e]]",
"command_line_arguments": "/home/magnus/.local/venv/ml-agents/bin/mlagents-learn --resume config/ppo/PyramidsRND.yaml --env training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids1 --no-graphics",
"mlagents_version": "0.31.0.dev0",
"mlagents_envs_version": "0.31.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "1.11.0+cu102",
"numpy_version": "1.21.2",
"end_time_seconds": "1688317031"
},
"total": 3357.0718518270005,
"count": 1,
"self": 0.3190375850026612,
"children": {
"run_training.setup": {
"total": 0.0074317609978606924,
"count": 1,
"self": 0.0074317609978606924
},
"TrainerController.start_learning": {
"total": 3356.745382481,
"count": 1,
"self": 2.9512564832948556,
"children": {
"TrainerController._reset_env": {
"total": 2.612207900998328,
"count": 1,
"self": 2.612207900998328
},
"TrainerController.advance": {
"total": 3351.123781086706,
"count": 198806,
"self": 2.758197276325518,
"children": {
"env_step": {
"total": 2284.4431627157064,
"count": 198806,
"self": 2075.268968008055,
"children": {
"SubprocessEnvManager._take_step": {
"total": 207.21744262683205,
"count": 198806,
"self": 8.163825870218716,
"children": {
"TorchPolicy.evaluate": {
"total": 199.05361675661334,
"count": 187530,
"self": 199.05361675661334
}
}
},
"workers": {
"total": 1.9567520808195695,
"count": 198806,
"self": 0.0,
"children": {
"worker_root": {
"total": 3353.061701289891,
"count": 198806,
"is_parallel": true,
"self": 1475.1090512383926,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0010249199985992163,
"count": 1,
"is_parallel": true,
"self": 0.0002954640040115919,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0007294559945876244,
"count": 8,
"is_parallel": true,
"self": 0.0007294559945876244
}
}
},
"UnityEnvironment.step": {
"total": 0.022472407999885036,
"count": 1,
"is_parallel": true,
"self": 0.0005902970005990937,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00018204400112153962,
"count": 1,
"is_parallel": true,
"self": 0.00018204400112153962
},
"communicator.exchange": {
"total": 0.020123067999520572,
"count": 1,
"is_parallel": true,
"self": 0.020123067999520572
},
"steps_from_proto": {
"total": 0.0015769989986438304,
"count": 1,
"is_parallel": true,
"self": 0.00024784100605756976,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0013291579925862607,
"count": 8,
"is_parallel": true,
"self": 0.0013291579925862607
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1877.9526500514985,
"count": 198805,
"is_parallel": true,
"self": 60.930162151478726,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 31.088286421967496,
"count": 198805,
"is_parallel": true,
"self": 31.088286421967496
},
"communicator.exchange": {
"total": 1619.3520383784635,
"count": 198805,
"is_parallel": true,
"self": 1619.3520383784635
},
"steps_from_proto": {
"total": 166.5821630995888,
"count": 198805,
"is_parallel": true,
"self": 31.637742599959893,
"children": {
"_process_rank_one_or_two_observation": {
"total": 134.9444204996289,
"count": 1590440,
"is_parallel": true,
"self": 134.9444204996289
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 1063.922421094674,
"count": 198806,
"self": 5.016832006564073,
"children": {
"process_trajectory": {
"total": 198.75968678008576,
"count": 198806,
"self": 198.33774820808685,
"children": {
"RLTrainer._checkpoint": {
"total": 0.4219385719989077,
"count": 6,
"self": 0.4219385719989077
}
}
},
"_update_policy": {
"total": 860.1459023080242,
"count": 1421,
"self": 555.9807385543063,
"children": {
"TorchPPOOptimizer.update": {
"total": 304.16516375371793,
"count": 68340,
"self": 304.16516375371793
}
}
}
}
}
}
},
"trainer_threads": {
"total": 6.219997885636985e-07,
"count": 1,
"self": 6.219997885636985e-07
},
"TrainerController._save_models": {
"total": 0.05813638800100307,
"count": 1,
"self": 0.0008321550012624357,
"children": {
"RLTrainer._checkpoint": {
"total": 0.057304232999740634,
"count": 1,
"self": 0.057304232999740634
}
}
}
}
}
}
}