ppo-Pyramids / run_logs /timers.json
Scrwed's picture
First test with updates
bdffb7b
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.24896404147148132,
"min": 0.24896404147148132,
"max": 1.3713804483413696,
"count": 25
},
"Pyramids.Policy.Entropy.sum": {
"value": 12424.3017578125,
"min": 12424.3017578125,
"max": 68634.8515625,
"count": 25
},
"Pyramids.Step.mean": {
"value": 1249934.0,
"min": 49920.0,
"max": 1249934.0,
"count": 25
},
"Pyramids.Step.sum": {
"value": 1249934.0,
"min": 49920.0,
"max": 1249934.0,
"count": 25
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.6520876884460449,
"min": -0.07605277746915817,
"max": 0.747703492641449,
"count": 25
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 307.13330078125,
"min": -30.57321548461914,
"max": 367.8701171875,
"count": 25
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.00311981956474483,
"min": -0.002151226857677102,
"max": 0.40616461634635925,
"count": 25
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 1.4694349765777588,
"min": -0.9874131679534912,
"max": 162.05967712402344,
"count": 25
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.07121741160180123,
"min": 0.0667192919373525,
"max": 0.07297704907692235,
"count": 25
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 1.6380004668414283,
"min": 0.8543848719599392,
"max": 1.7137005866331723,
"count": 25
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.013248737996774795,
"min": 0.0009852832274306305,
"max": 0.01500675338067357,
"count": 25
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.30472097392582026,
"min": 0.02069094777604324,
"max": 0.35135184237636474,
"count": 25
},
"Pyramids.Policy.LearningRate.mean": {
"value": 1.000011539130088e-05,
"min": 1.000011539130088e-05,
"max": 0.0004895424020915201,
"count": 25
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00023000265399992023,
"min": 0.00023000265399992023,
"max": 0.009436869212626238,
"count": 25
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10200000347826087,
"min": 0.10200000347826087,
"max": 0.19790848000000003,
"count": 25
},
"Pyramids.Policy.Epsilon.sum": {
"value": 2.34600008,
"min": 2.34600008,
"max": 4.18737376,
"count": 25
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0002098003474782611,
"min": 0.0002098003474782611,
"max": 0.009791057152,
"count": 25
},
"Pyramids.Policy.Beta.sum": {
"value": 0.004825407992000005,
"min": 0.004825407992000005,
"max": 0.18877863862400002,
"count": 25
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.00874233152717352,
"min": 0.00874233152717352,
"max": 0.20817802846431732,
"count": 25
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.20107361674308777,
"min": 0.20107361674308777,
"max": 2.498136281967163,
"count": 25
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 296.2831325301205,
"min": 253.17326732673268,
"max": 999.0,
"count": 25
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 49183.0,
"min": 47579.0,
"max": 52903.0,
"count": 25
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.5964502828742215,
"min": -1.0000000521540642,
"max": 1.7266706309063518,
"count": 25
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 266.607197239995,
"min": -48.00000250339508,
"max": 347.0607968121767,
"count": 25
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.5964502828742215,
"min": -1.0000000521540642,
"max": 1.7266706309063518,
"count": 25
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 266.607197239995,
"min": -48.00000250339508,
"max": 347.0607968121767,
"count": 25
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.027748993725406457,
"min": 0.024853757902529826,
"max": 4.531360094901174,
"count": 25
},
"Pyramids.Policy.RndReward.sum": {
"value": 4.634081952142878,
"min": 4.634081952142878,
"max": 217.50528455525637,
"count": 25
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 25
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 25
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1673852303",
"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-Trainging2 --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": "1673854853"
},
"total": 2550.227713464,
"count": 1,
"self": 0.42434544199932134,
"children": {
"run_training.setup": {
"total": 0.10303959800012308,
"count": 1,
"self": 0.10303959800012308
},
"TrainerController.start_learning": {
"total": 2549.7003284240004,
"count": 1,
"self": 1.4452616859216505,
"children": {
"TrainerController._reset_env": {
"total": 6.202183891000004,
"count": 1,
"self": 6.202183891000004
},
"TrainerController.advance": {
"total": 2541.927548809079,
"count": 80197,
"self": 1.5209907230209865,
"children": {
"env_step": {
"total": 1716.3584119130305,
"count": 80197,
"self": 1589.1763365150846,
"children": {
"SubprocessEnvManager._take_step": {
"total": 126.29957709800851,
"count": 80197,
"self": 5.143962063047184,
"children": {
"TorchPolicy.evaluate": {
"total": 121.15561503496133,
"count": 78186,
"self": 39.82934110292558,
"children": {
"TorchPolicy.sample_actions": {
"total": 81.32627393203575,
"count": 78186,
"self": 81.32627393203575
}
}
}
}
},
"workers": {
"total": 0.8824982999374242,
"count": 80197,
"self": 0.0,
"children": {
"worker_root": {
"total": 2544.4011875560404,
"count": 80197,
"is_parallel": true,
"self": 1073.2867285961224,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0017514840001240373,
"count": 1,
"is_parallel": true,
"self": 0.0006411189997379552,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001110365000386082,
"count": 8,
"is_parallel": true,
"self": 0.001110365000386082
}
}
},
"UnityEnvironment.step": {
"total": 0.06788392299995394,
"count": 1,
"is_parallel": true,
"self": 0.0005245350002951454,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0004341179999300948,
"count": 1,
"is_parallel": true,
"self": 0.0004341179999300948
},
"communicator.exchange": {
"total": 0.0653619949998756,
"count": 1,
"is_parallel": true,
"self": 0.0653619949998756
},
"steps_from_proto": {
"total": 0.0015632749998530926,
"count": 1,
"is_parallel": true,
"self": 0.00039875899983599083,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011645160000171018,
"count": 8,
"is_parallel": true,
"self": 0.0011645160000171018
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1471.114458959918,
"count": 80196,
"is_parallel": true,
"self": 33.920529114921464,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 26.70582958996897,
"count": 80196,
"is_parallel": true,
"self": 26.70582958996897
},
"communicator.exchange": {
"total": 1288.1845110940183,
"count": 80196,
"is_parallel": true,
"self": 1288.1845110940183
},
"steps_from_proto": {
"total": 122.30358916100931,
"count": 80196,
"is_parallel": true,
"self": 26.41566861006436,
"children": {
"_process_rank_one_or_two_observation": {
"total": 95.88792055094495,
"count": 641568,
"is_parallel": true,
"self": 95.88792055094495
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 824.0481461730274,
"count": 80197,
"self": 2.7291802370398273,
"children": {
"process_trajectory": {
"total": 177.53446862199053,
"count": 80197,
"self": 177.2787857959902,
"children": {
"RLTrainer._checkpoint": {
"total": 0.25568282600033854,
"count": 2,
"self": 0.25568282600033854
}
}
},
"_update_policy": {
"total": 643.784497313997,
"count": 564,
"self": 239.95744739697443,
"children": {
"TorchPPOOptimizer.update": {
"total": 403.8270499170226,
"count": 28503,
"self": 403.8270499170226
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.27099974633893e-06,
"count": 1,
"self": 1.27099974633893e-06
},
"TrainerController._save_models": {
"total": 0.1253327670001454,
"count": 1,
"self": 0.002084943000227213,
"children": {
"RLTrainer._checkpoint": {
"total": 0.12324782399991818,
"count": 1,
"self": 0.12324782399991818
}
}
}
}
}
}
}