ppo-Pyramid / run_logs /timers.json
jyesr's picture
First Push
bd232cf verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.5650244355201721,
"min": 0.5650244355201721,
"max": 1.4269523620605469,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 16932.65234375,
"min": 16932.65234375,
"max": 43288.02734375,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989889.0,
"min": 29952.0,
"max": 989889.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989889.0,
"min": 29952.0,
"max": 989889.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.2928575575351715,
"min": -0.10202005505561829,
"max": 0.2928575575351715,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 75.85010528564453,
"min": -24.586833953857422,
"max": 75.85010528564453,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.010896404273808002,
"min": -0.025064747780561447,
"max": 0.3033257722854614,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 2.8221688270568848,
"min": -6.491769790649414,
"max": 72.79818725585938,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06843657581235416,
"min": 0.06351763047236472,
"max": 0.07418307405001087,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9581120613729581,
"min": 0.4911929619426293,
"max": 1.0488071050631869,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.012636160040398994,
"min": 0.0005076472885833226,
"max": 0.012636160040398994,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.1769062405655859,
"min": 0.0040611783086665805,
"max": 0.1769062405655859,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.628990314178571e-06,
"min": 7.628990314178571e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.0001068058643985,
"min": 0.0001068058643985,
"max": 0.0033750820749726997,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.1025429642857143,
"min": 0.1025429642857143,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4356015000000002,
"min": 1.3691136000000002,
"max": 2.4250273,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0002640421321428572,
"min": 0.0002640421321428572,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.00369658985,
"min": 0.00369658985,
"max": 0.11252022727,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.008489670231938362,
"min": 0.008489670231938362,
"max": 0.4667523503303528,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.11885537952184677,
"min": 0.11885537952184677,
"max": 3.2672665119171143,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 543.2363636363636,
"min": 543.2363636363636,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 29878.0,
"min": 15984.0,
"max": 32395.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.165781798281453,
"min": -1.0000000521540642,
"max": 1.165781798281453,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 64.11799890547991,
"min": -32.000001668930054,
"max": 64.11799890547991,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.165781798281453,
"min": -1.0000000521540642,
"max": 1.165781798281453,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 64.11799890547991,
"min": -32.000001668930054,
"max": 64.11799890547991,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.04705585517909971,
"min": 0.04705585517909971,
"max": 10.455010060220957,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 2.5880720348504838,
"min": 2.5880720348504838,
"max": 167.2801609635353,
"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": "1708933743",
"python_version": "3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]",
"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 --force",
"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": "1708936102"
},
"total": 2358.404444975,
"count": 1,
"self": 0.5369859100001122,
"children": {
"run_training.setup": {
"total": 0.050553755000009915,
"count": 1,
"self": 0.050553755000009915
},
"TrainerController.start_learning": {
"total": 2357.81690531,
"count": 1,
"self": 1.7188306190132607,
"children": {
"TrainerController._reset_env": {
"total": 2.522374769999942,
"count": 1,
"self": 2.522374769999942
},
"TrainerController.advance": {
"total": 2353.4787344529873,
"count": 63315,
"self": 1.7743434109311238,
"children": {
"env_step": {
"total": 1693.1593569209954,
"count": 63315,
"self": 1538.089040188952,
"children": {
"SubprocessEnvManager._take_step": {
"total": 154.01399924903103,
"count": 63315,
"self": 5.481878953037267,
"children": {
"TorchPolicy.evaluate": {
"total": 148.53212029599376,
"count": 62556,
"self": 148.53212029599376
}
}
},
"workers": {
"total": 1.0563174830123216,
"count": 63315,
"self": 0.0,
"children": {
"worker_root": {
"total": 2351.9709405419767,
"count": 63315,
"is_parallel": true,
"self": 950.1568276580042,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0030262289999427594,
"count": 1,
"is_parallel": true,
"self": 0.0008911829997941823,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.002135046000148577,
"count": 8,
"is_parallel": true,
"self": 0.002135046000148577
}
}
},
"UnityEnvironment.step": {
"total": 0.05711964500005706,
"count": 1,
"is_parallel": true,
"self": 0.0006721400000060385,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0005888220000542788,
"count": 1,
"is_parallel": true,
"self": 0.0005888220000542788
},
"communicator.exchange": {
"total": 0.05395298600001297,
"count": 1,
"is_parallel": true,
"self": 0.05395298600001297
},
"steps_from_proto": {
"total": 0.0019056969999837747,
"count": 1,
"is_parallel": true,
"self": 0.00039753600003678,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0015081609999469947,
"count": 8,
"is_parallel": true,
"self": 0.0015081609999469947
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1401.8141128839725,
"count": 63314,
"is_parallel": true,
"self": 38.87951073996078,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 27.683347618006565,
"count": 63314,
"is_parallel": true,
"self": 27.683347618006565
},
"communicator.exchange": {
"total": 1220.2686652730445,
"count": 63314,
"is_parallel": true,
"self": 1220.2686652730445
},
"steps_from_proto": {
"total": 114.98258925296068,
"count": 63314,
"is_parallel": true,
"self": 24.108969811953784,
"children": {
"_process_rank_one_or_two_observation": {
"total": 90.8736194410069,
"count": 506512,
"is_parallel": true,
"self": 90.8736194410069
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 658.5450341210609,
"count": 63315,
"self": 3.037311855069447,
"children": {
"process_trajectory": {
"total": 135.38670211699196,
"count": 63315,
"self": 135.1506770879921,
"children": {
"RLTrainer._checkpoint": {
"total": 0.23602502899984756,
"count": 2,
"self": 0.23602502899984756
}
}
},
"_update_policy": {
"total": 520.1210201489995,
"count": 439,
"self": 304.392472698019,
"children": {
"TorchPPOOptimizer.update": {
"total": 215.7285474509805,
"count": 22827,
"self": 215.7285474509805
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.03899992609513e-06,
"count": 1,
"self": 1.03899992609513e-06
},
"TrainerController._save_models": {
"total": 0.09696442899985414,
"count": 1,
"self": 0.002382489999945392,
"children": {
"RLTrainer._checkpoint": {
"total": 0.09458193899990874,
"count": 1,
"self": 0.09458193899990874
}
}
}
}
}
}
}