SungjunEom's picture
hi
6aac0a6 verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.508629322052002,
"min": 0.49945271015167236,
"max": 1.4111984968185425,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 15340.259765625,
"min": 14943.625,
"max": 42810.1171875,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989998.0,
"min": 29959.0,
"max": 989998.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989998.0,
"min": 29959.0,
"max": 989998.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.2774900197982788,
"min": -0.1050730049610138,
"max": 0.4180706739425659,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 71.59242248535156,
"min": -25.21752166748047,
"max": 112.46101379394531,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.03267830237746239,
"min": 0.01298768725246191,
"max": 0.5670977830886841,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 8.431001663208008,
"min": 3.207958698272705,
"max": 134.96926879882812,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06757368302047423,
"min": 0.06322578654163392,
"max": 0.07281671626490999,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9460315622866392,
"min": 0.47932300141068496,
"max": 1.0547387134375088,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.013799651446195039,
"min": 0.00028901319463792385,
"max": 0.03074120846068628,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.19319512024673055,
"min": 0.003468158335655086,
"max": 0.2487676479478764,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.387040394828571e-06,
"min": 7.387040394828571e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.0001034185655276,
"min": 0.0001034185655276,
"max": 0.0036090099969967,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10246231428571428,
"min": 0.10246231428571428,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4344724,
"min": 1.3886848,
"max": 2.5030033000000005,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00025598519714285724,
"min": 0.00025598519714285724,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0035837927600000014,
"min": 0.0035837927600000014,
"max": 0.12031002967000001,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.015394777059555054,
"min": 0.015394777059555054,
"max": 0.7876579165458679,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.21552687883377075,
"min": 0.21552687883377075,
"max": 5.51360559463501,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 523.3728813559322,
"min": 392.67123287671234,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 30879.0,
"min": 16758.0,
"max": 34882.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.069725397027145,
"min": -0.9999750526621938,
"max": 1.4154739491131207,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 63.113798424601555,
"min": -31.9992016851902,
"max": 103.32959828525782,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.069725397027145,
"min": -0.9999750526621938,
"max": 1.4154739491131207,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 63.113798424601555,
"min": -31.9992016851902,
"max": 103.32959828525782,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.08403618890896708,
"min": 0.06956506816053137,
"max": 16.439932984245175,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 4.958135145629058,
"min": 4.755764963570982,
"max": 279.47886073216796,
"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": "1707461799",
"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/Pyramids2/Pyramids/Pyramids --run-id=Pyramids Training 3 --no-graphics",
"mlagents_version": "1.1.0.dev0",
"mlagents_envs_version": "1.1.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.2.0+cu121",
"numpy_version": "1.23.5",
"end_time_seconds": "1707463878"
},
"total": 2079.736673292,
"count": 1,
"self": 0.4911426879998544,
"children": {
"run_training.setup": {
"total": 0.0486979990000691,
"count": 1,
"self": 0.0486979990000691
},
"TrainerController.start_learning": {
"total": 2079.196832605,
"count": 1,
"self": 1.4001195610662762,
"children": {
"TrainerController._reset_env": {
"total": 2.3649198140001317,
"count": 1,
"self": 2.3649198140001317
},
"TrainerController.advance": {
"total": 2075.3455833429343,
"count": 63420,
"self": 1.5330758721020175,
"children": {
"env_step": {
"total": 1460.5596965688828,
"count": 63420,
"self": 1330.9461030777993,
"children": {
"SubprocessEnvManager._take_step": {
"total": 128.76203565308333,
"count": 63420,
"self": 4.504562303978673,
"children": {
"TorchPolicy.evaluate": {
"total": 124.25747334910466,
"count": 62558,
"self": 124.25747334910466
}
}
},
"workers": {
"total": 0.8515578380001898,
"count": 63420,
"self": 0.0,
"children": {
"worker_root": {
"total": 2074.054900891995,
"count": 63420,
"is_parallel": true,
"self": 858.4091778369557,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.001986274999580928,
"count": 1,
"is_parallel": true,
"self": 0.0006675990007352084,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0013186759988457197,
"count": 8,
"is_parallel": true,
"self": 0.0013186759988457197
}
}
},
"UnityEnvironment.step": {
"total": 0.04892965399994864,
"count": 1,
"is_parallel": true,
"self": 0.0005986959995425423,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00046777900024608243,
"count": 1,
"is_parallel": true,
"self": 0.00046777900024608243
},
"communicator.exchange": {
"total": 0.04625908700018044,
"count": 1,
"is_parallel": true,
"self": 0.04625908700018044
},
"steps_from_proto": {
"total": 0.0016040919999795733,
"count": 1,
"is_parallel": true,
"self": 0.00032817699957377044,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0012759150004058029,
"count": 8,
"is_parallel": true,
"self": 0.0012759150004058029
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1215.6457230550395,
"count": 63419,
"is_parallel": true,
"self": 34.835662975116975,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 23.55985296193012,
"count": 63419,
"is_parallel": true,
"self": 23.55985296193012
},
"communicator.exchange": {
"total": 1059.9628566199908,
"count": 63419,
"is_parallel": true,
"self": 1059.9628566199908
},
"steps_from_proto": {
"total": 97.28735049800162,
"count": 63419,
"is_parallel": true,
"self": 19.321935031180146,
"children": {
"_process_rank_one_or_two_observation": {
"total": 77.96541546682147,
"count": 507352,
"is_parallel": true,
"self": 77.96541546682147
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 613.2528109019495,
"count": 63420,
"self": 2.688237308925636,
"children": {
"process_trajectory": {
"total": 121.14304332802385,
"count": 63420,
"self": 120.90128575102426,
"children": {
"RLTrainer._checkpoint": {
"total": 0.24175757699958922,
"count": 2,
"self": 0.24175757699958922
}
}
},
"_update_policy": {
"total": 489.421530265,
"count": 452,
"self": 288.16299733104916,
"children": {
"TorchPPOOptimizer.update": {
"total": 201.25853293395085,
"count": 22869,
"self": 201.25853293395085
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.3919998309575021e-06,
"count": 1,
"self": 1.3919998309575021e-06
},
"TrainerController._save_models": {
"total": 0.08620849499948235,
"count": 1,
"self": 0.0014162489987938898,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08479224600068846,
"count": 1,
"self": 0.08479224600068846
}
}
}
}
}
}
}