ppo-Pyramids / run_logs /timers.json
Changyu Gao
First Push
b26adec
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.925483763217926,
"min": 0.925483763217926,
"max": 1.4854224920272827,
"count": 5
},
"Pyramids.Policy.Entropy.sum": {
"value": 28105.091796875,
"min": 28105.091796875,
"max": 45061.77734375,
"count": 5
},
"Pyramids.Step.mean": {
"value": 149982.0,
"min": 29952.0,
"max": 149982.0,
"count": 5
},
"Pyramids.Step.sum": {
"value": 149982.0,
"min": 29952.0,
"max": 149982.0,
"count": 5
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": -0.05815637856721878,
"min": -0.09418731182813644,
"max": 0.0001792570692487061,
"count": 5
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": -14.131999969482422,
"min": -22.699142456054688,
"max": 0.04248392581939697,
"count": 5
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.08950106799602509,
"min": 0.08950106799602509,
"max": 0.3058807849884033,
"count": 5
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 21.748760223388672,
"min": 21.748760223388672,
"max": 72.49374389648438,
"count": 5
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06793653849306233,
"min": 0.06793653849306233,
"max": 0.0716153584697478,
"count": 5
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.8831750004098102,
"min": 0.4887146136687329,
"max": 0.8831750004098102,
"count": 5
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.002622824082214735,
"min": 0.0005557609182493178,
"max": 0.006334852572088312,
"count": 5
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.034096713068791557,
"min": 0.0044460873459945425,
"max": 0.044343968004618185,
"count": 5
},
"Pyramids.Policy.LearningRate.mean": {
"value": 0.00021874271939346154,
"min": 0.00021874271939346154,
"max": 0.00029030126037577137,
"count": 5
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.002843655352115,
"min": 0.0019146816617728003,
"max": 0.002843655352115,
"count": 5
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.1729142307692308,
"min": 0.1729142307692308,
"max": 0.19676708571428575,
"count": 5
},
"Pyramids.Policy.Epsilon.sum": {
"value": 2.247885,
"min": 1.3382272,
"max": 2.247885,
"count": 5
},
"Pyramids.Policy.Beta.mean": {
"value": 0.007294131653846152,
"min": 0.007294131653846152,
"max": 0.00967703186285714,
"count": 5
},
"Pyramids.Policy.Beta.sum": {
"value": 0.09482371149999998,
"min": 0.06382889727999999,
"max": 0.09482371149999998,
"count": 5
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.07258495688438416,
"min": 0.07258495688438416,
"max": 0.4456627666950226,
"count": 5
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.9436044692993164,
"min": 0.9436044692993164,
"max": 3.1196393966674805,
"count": 5
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 951.2857142857143,
"min": 942.8787878787879,
"max": 999.0,
"count": 5
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 33295.0,
"min": 15984.0,
"max": 33295.0,
"count": 5
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": -0.6663200523172106,
"min": -1.0000000521540642,
"max": -0.6406727760578647,
"count": 5
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": -23.32120183110237,
"min": -32.000001668930054,
"max": -16.000000834465027,
"count": 5
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": -0.6663200523172106,
"min": -1.0000000521540642,
"max": -0.6406727760578647,
"count": 5
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": -23.32120183110237,
"min": -32.000001668930054,
"max": -16.000000834465027,
"count": 5
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.7318932928144932,
"min": 0.7318932928144932,
"max": 9.054668163880706,
"count": 5
},
"Pyramids.Policy.RndReward.sum": {
"value": 25.61626524850726,
"min": 25.61626524850726,
"max": 144.8746906220913,
"count": 5
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 5
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 5
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1686166175",
"python_version": "3.10.11 (main, Apr 5 2023, 14:15:10) [GCC 9.4.0]",
"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",
"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": "1686166530"
},
"total": 355.3194051390003,
"count": 1,
"self": 0.4724490980004248,
"children": {
"run_training.setup": {
"total": 0.05494175700005144,
"count": 1,
"self": 0.05494175700005144
},
"TrainerController.start_learning": {
"total": 354.79201428399983,
"count": 1,
"self": 0.23012636901921724,
"children": {
"TrainerController._reset_env": {
"total": 4.8237584920000245,
"count": 1,
"self": 4.8237584920000245
},
"TrainerController.advance": {
"total": 349.7348349659808,
"count": 11205,
"self": 0.23454281000113042,
"children": {
"env_step": {
"total": 236.48923420098936,
"count": 11205,
"self": 216.9719004049955,
"children": {
"SubprocessEnvManager._take_step": {
"total": 19.382279875993618,
"count": 11205,
"self": 0.825565159010921,
"children": {
"TorchPolicy.evaluate": {
"total": 18.556714716982697,
"count": 11177,
"self": 18.556714716982697
}
}
},
"workers": {
"total": 0.13505392000024585,
"count": 11204,
"self": 0.0,
"children": {
"worker_root": {
"total": 354.09471588100655,
"count": 11204,
"is_parallel": true,
"self": 156.39775927301412,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.004157110000051034,
"count": 1,
"is_parallel": true,
"self": 0.002264487999582343,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0018926220004686911,
"count": 8,
"is_parallel": true,
"self": 0.0018926220004686911
}
}
},
"UnityEnvironment.step": {
"total": 0.04948453800011521,
"count": 1,
"is_parallel": true,
"self": 0.0005666399999881833,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0005429540001387068,
"count": 1,
"is_parallel": true,
"self": 0.0005429540001387068
},
"communicator.exchange": {
"total": 0.046561502000031396,
"count": 1,
"is_parallel": true,
"self": 0.046561502000031396
},
"steps_from_proto": {
"total": 0.0018134419999569218,
"count": 1,
"is_parallel": true,
"self": 0.00035785599993687356,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0014555860000200482,
"count": 8,
"is_parallel": true,
"self": 0.0014555860000200482
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 197.69695660799243,
"count": 11203,
"is_parallel": true,
"self": 5.61437618796117,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 4.01434996302919,
"count": 11203,
"is_parallel": true,
"self": 4.01434996302919
},
"communicator.exchange": {
"total": 170.770776979007,
"count": 11203,
"is_parallel": true,
"self": 170.770776979007
},
"steps_from_proto": {
"total": 17.297453477995077,
"count": 11203,
"is_parallel": true,
"self": 3.5061899969916794,
"children": {
"_process_rank_one_or_two_observation": {
"total": 13.791263481003398,
"count": 89624,
"is_parallel": true,
"self": 13.791263481003398
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 113.01105795499029,
"count": 11204,
"self": 0.3113298669898086,
"children": {
"process_trajectory": {
"total": 17.913187625001456,
"count": 11204,
"self": 17.913187625001456
},
"_update_policy": {
"total": 94.78654046299903,
"count": 58,
"self": 60.67647631799355,
"children": {
"TorchPPOOptimizer.update": {
"total": 34.11006414500548,
"count": 4122,
"self": 34.11006414500548
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.3730000318901148e-06,
"count": 1,
"self": 1.3730000318901148e-06
},
"TrainerController._save_models": {
"total": 0.0032930839997789008,
"count": 1,
"self": 2.256399966427125e-05,
"children": {
"RLTrainer._checkpoint": {
"total": 0.0032705200001146295,
"count": 1,
"self": 0.0032705200001146295
}
}
}
}
}
}
}