ppo-PyramidsRND / run_logs /timers.json
letaldir's picture
Second Push
aea10fd verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.18347740173339844,
"min": 0.16536884009838104,
"max": 0.3051706850528717,
"count": 20
},
"Pyramids.Policy.Entropy.sum": {
"value": 5539.5498046875,
"min": 4908.14697265625,
"max": 9409.0224609375,
"count": 20
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 257.2357723577236,
"min": 229.29032258064515,
"max": 286.1926605504587,
"count": 20
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 31640.0,
"min": 28432.0,
"max": 31640.0,
"count": 20
},
"Pyramids.Step.mean": {
"value": 1799967.0,
"min": 1229987.0,
"max": 1799967.0,
"count": 20
},
"Pyramids.Step.sum": {
"value": 1799967.0,
"min": 1229987.0,
"max": 1799967.0,
"count": 20
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.7283996939659119,
"min": 0.6516091227531433,
"max": 0.7944933772087097,
"count": 20
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 212.6927032470703,
"min": 187.01181030273438,
"max": 236.759033203125,
"count": 20
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": -0.01562570407986641,
"min": -0.01562570407986641,
"max": 0.047894831746816635,
"count": 20
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": -4.5627055168151855,
"min": -4.5627055168151855,
"max": 13.841606140136719,
"count": 20
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.6614390073752985,
"min": 1.6528878328956176,
"max": 1.7657983686414458,
"count": 20
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 204.3569979071617,
"min": 175.7153990715742,
"max": 221.81499876081944,
"count": 20
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.6614390073752985,
"min": 1.6528878328956176,
"max": 1.7657983686414458,
"count": 20
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 204.3569979071617,
"min": 175.7153990715742,
"max": 221.81499876081944,
"count": 20
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.021665721189995425,
"min": 0.020442040216332603,
"max": 0.026402303401928808,
"count": 20
},
"Pyramids.Policy.RndReward.sum": {
"value": 2.6648837063694373,
"min": 2.473486866176245,
"max": 2.960366822386277,
"count": 20
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.0696433908888139,
"min": 0.06593222917426549,
"max": 0.07108483445370525,
"count": 20
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9750074724433945,
"min": 0.9230512084397169,
"max": 1.056874103579378,
"count": 20
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.015932606914984285,
"min": 0.012798120494283336,
"max": 0.016656490289617566,
"count": 20
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.22305649680977999,
"min": 0.1791736869199667,
"max": 0.23319086405464592,
"count": 20
},
"Pyramids.Policy.LearningRate.mean": {
"value": 2.4778848883571454e-06,
"min": 2.4778848883571454e-06,
"max": 9.736857944859126e-05,
"count": 20
},
"Pyramids.Policy.LearningRate.sum": {
"value": 3.4690388437000035e-05,
"min": 3.4690388437000035e-05,
"max": 0.0013631601122802777,
"count": 20
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10082592857142857,
"min": 0.10082592857142857,
"max": 0.13245617063492066,
"count": 20
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.411563,
"min": 1.411563,
"max": 1.8625565555555554,
"count": 20
},
"Pyramids.Policy.Beta.mean": {
"value": 9.251026428571435e-05,
"min": 9.251026428571435e-05,
"max": 0.003252371446428571,
"count": 20
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0012951437000000008,
"min": 0.0012951437000000008,
"max": 0.045533200249999996,
"count": 20
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.008193425834178925,
"min": 0.008193425834178925,
"max": 0.009500919841229916,
"count": 20
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.11470796167850494,
"min": 0.11470796167850494,
"max": 0.13423441350460052,
"count": 20
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 20
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 20
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1762951087",
"python_version": "3.10.12 (main, Jul 5 2023, 18:54:27) [GCC 11.2.0]",
"command_line_arguments": "/usr/local/bin/mlagents-learn ./config/ppo/PyramidsRND.yaml --resume --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids Training --no-graphics",
"mlagents_version": "1.2.0.dev0",
"mlagents_envs_version": "1.2.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.8.0+cu128",
"numpy_version": "1.23.5",
"end_time_seconds": "1762952568"
},
"total": 1480.701929109,
"count": 1,
"self": 0.532957625000563,
"children": {
"run_training.setup": {
"total": 0.023814800999389263,
"count": 1,
"self": 0.023814800999389263
},
"TrainerController.start_learning": {
"total": 1480.145156683,
"count": 1,
"self": 0.7973223673343455,
"children": {
"TrainerController._reset_env": {
"total": 2.084303745001307,
"count": 1,
"self": 2.084303745001307
},
"TrainerController.advance": {
"total": 1477.1874179816623,
"count": 39419,
"self": 0.8337124696463434,
"children": {
"env_step": {
"total": 1089.8707667620656,
"count": 39419,
"self": 1002.9535249321434,
"children": {
"SubprocessEnvManager._take_step": {
"total": 86.43570298192208,
"count": 39419,
"self": 2.7428888748327154,
"children": {
"TorchPolicy.evaluate": {
"total": 83.69281410708936,
"count": 37559,
"self": 83.69281410708936
}
}
},
"workers": {
"total": 0.4815388480001275,
"count": 39419,
"self": 0.0,
"children": {
"worker_root": {
"total": 1476.2343495521945,
"count": 39419,
"is_parallel": true,
"self": 542.2397924860525,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.001774239999576821,
"count": 1,
"is_parallel": true,
"self": 0.000596388999838382,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001177850999738439,
"count": 8,
"is_parallel": true,
"self": 0.001177850999738439
}
}
},
"UnityEnvironment.step": {
"total": 0.05186884599970654,
"count": 1,
"is_parallel": true,
"self": 0.0005691830010619015,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0004935919987474335,
"count": 1,
"is_parallel": true,
"self": 0.0004935919987474335
},
"communicator.exchange": {
"total": 0.04909174000022176,
"count": 1,
"is_parallel": true,
"self": 0.04909174000022176
},
"steps_from_proto": {
"total": 0.0017143309996754397,
"count": 1,
"is_parallel": true,
"self": 0.00032570099938311614,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0013886300002923235,
"count": 8,
"is_parallel": true,
"self": 0.0013886300002923235
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 933.9945570661421,
"count": 39418,
"is_parallel": true,
"self": 19.88177738012928,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 13.84117243197943,
"count": 39418,
"is_parallel": true,
"self": 13.84117243197943
},
"communicator.exchange": {
"total": 835.4911941160371,
"count": 39418,
"is_parallel": true,
"self": 835.4911941160371
},
"steps_from_proto": {
"total": 64.78041313799622,
"count": 39418,
"is_parallel": true,
"self": 13.512100370020562,
"children": {
"_process_rank_one_or_two_observation": {
"total": 51.26831276797566,
"count": 315344,
"is_parallel": true,
"self": 51.26831276797566
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 386.48293874995034,
"count": 39419,
"self": 1.6931462749180355,
"children": {
"process_trajectory": {
"total": 77.74215061902578,
"count": 39419,
"self": 77.63888940402649,
"children": {
"RLTrainer._checkpoint": {
"total": 0.10326121499929286,
"count": 1,
"self": 0.10326121499929286
}
}
},
"_update_policy": {
"total": 307.0476418560065,
"count": 283,
"self": 171.16670905297906,
"children": {
"TorchPPOOptimizer.update": {
"total": 135.88093280302746,
"count": 13647,
"self": 135.88093280302746
}
}
}
}
}
}
},
"trainer_threads": {
"total": 8.880015229806304e-07,
"count": 1,
"self": 8.880015229806304e-07
},
"TrainerController._save_models": {
"total": 0.07611170100062736,
"count": 1,
"self": 0.0013987319998705061,
"children": {
"RLTrainer._checkpoint": {
"total": 0.07471296900075686,
"count": 1,
"self": 0.07471296900075686
}
}
}
}
}
}
}