ppo-Pyramids / run_logs /timers.json
jyunyilin's picture
First traning of Pyramids
9182f98 verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.014203900471329689,
"min": 0.011453866958618164,
"max": 0.8908319473266602,
"count": 20
},
"Pyramids.Policy.Entropy.sum": {
"value": 719.96728515625,
"min": 580.5736083984375,
"max": 27970.1796875,
"count": 20
},
"Pyramids.Step.mean": {
"value": 999899.0,
"min": 49934.0,
"max": 999899.0,
"count": 20
},
"Pyramids.Step.sum": {
"value": 999899.0,
"min": 49934.0,
"max": 999899.0,
"count": 20
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": -0.09875194728374481,
"min": -0.11406919360160828,
"max": -0.04936850816011429,
"count": 20
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": -19.750389099121094,
"min": -22.813838958740234,
"max": -8.056174278259277,
"count": 20
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.26081907749176025,
"min": 0.2394992560148239,
"max": 0.6021887063980103,
"count": 20
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 52.163818359375,
"min": 48.13935089111328,
"max": 86.98306274414062,
"count": 20
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.05254962694458503,
"min": 0.04264529009265566,
"max": 0.05254962694458503,
"count": 20
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.5780458963904354,
"min": 0.127935870277967,
"max": 0.5780458963904354,
"count": 20
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.0002617719697247572,
"min": 0.00024220746825061852,
"max": 0.004379402419484298,
"count": 20
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.0028794916669723294,
"min": 0.0026642821507568038,
"max": 0.04002989382357289,
"count": 20
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.81228830502727e-06,
"min": 7.81228830502727e-06,
"max": 0.0002889266036911334,
"count": 20
},
"Pyramids.Policy.LearningRate.sum": {
"value": 8.593517135529997e-05,
"min": 8.593517135529997e-05,
"max": 0.003149436150188,
"count": 20
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10260406363636365,
"min": 0.10260406363636365,
"max": 0.19630886666666666,
"count": 20
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.1286447000000002,
"min": 0.5889266,
"max": 2.249812,
"count": 20
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00027014595727272724,
"min": 0.00027014595727272724,
"max": 0.00963125578,
"count": 20
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0029716055299999996,
"min": 0.0029716055299999996,
"max": 0.1049962188,
"count": 20
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.2442188262939453,
"min": 0.23793675005435944,
"max": 0.37168848514556885,
"count": 20
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 2.6864070892333984,
"min": 1.1150654554367065,
"max": 3.189399003982544,
"count": 20
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 999.0,
"min": 956.0588235294117,
"max": 999.0,
"count": 20
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 47952.0,
"min": 16253.0,
"max": 60475.0,
"count": 20
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": -0.9997875560075045,
"min": -0.9998400568962097,
"max": -0.7159793623562517,
"count": 20
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": -47.989802688360214,
"min": -52.98980298638344,
"max": -14.269000977277756,
"count": 20
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": -0.9997875560075045,
"min": -0.9998400568962097,
"max": -0.7159793623562517,
"count": 20
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": -47.989802688360214,
"min": -52.98980298638344,
"max": -14.269000977277756,
"count": 20
},
"Pyramids.Policy.RndReward.mean": {
"value": 2.4803407109187297,
"min": 2.391638230730094,
"max": 4.593142227214925,
"count": 20
},
"Pyramids.Policy.RndReward.sum": {
"value": 119.05635412409902,
"min": 78.08341786265373,
"max": 161.26219326257706,
"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": "1742183501",
"python_version": "3.10.12 (main, Jul 5 2023, 18:54:27) [GCC 11.2.0]",
"command_line_arguments": "/usr/local/bin/mlagents-learn /content/ml-agents/config/ppo/PyramidsRND.yaml --env=/content/ml-agents/training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids Training --resume --no-graphics",
"mlagents_version": "1.2.0.dev0",
"mlagents_envs_version": "1.2.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.6.0+cu124",
"numpy_version": "1.23.5",
"end_time_seconds": "1742185379"
},
"total": 1878.428894729,
"count": 1,
"self": 0.4765186150000318,
"children": {
"run_training.setup": {
"total": 0.024628646999872217,
"count": 1,
"self": 0.024628646999872217
},
"TrainerController.start_learning": {
"total": 1877.927747467,
"count": 1,
"self": 1.2897568870109808,
"children": {
"TrainerController._reset_env": {
"total": 2.3837820300000203,
"count": 1,
"self": 2.3837820300000203
},
"TrainerController.advance": {
"total": 1874.1523743169892,
"count": 61701,
"self": 1.4779112529663507,
"children": {
"env_step": {
"total": 1337.2723673840571,
"count": 61701,
"self": 1168.6752160041176,
"children": {
"SubprocessEnvManager._take_step": {
"total": 167.7450551679724,
"count": 61701,
"self": 5.116065561934192,
"children": {
"TorchPolicy.evaluate": {
"total": 162.6289896060382,
"count": 61256,
"self": 162.6289896060382
}
}
},
"workers": {
"total": 0.8520962119671367,
"count": 61701,
"self": 0.0,
"children": {
"worker_root": {
"total": 1872.681595069047,
"count": 61701,
"is_parallel": true,
"self": 820.9112825771165,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.001929770000060671,
"count": 1,
"is_parallel": true,
"self": 0.000639861999616187,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001289908000444484,
"count": 8,
"is_parallel": true,
"self": 0.001289908000444484
}
}
},
"UnityEnvironment.step": {
"total": 0.09287251099999594,
"count": 1,
"is_parallel": true,
"self": 0.0005381920000218088,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.000522244999956456,
"count": 1,
"is_parallel": true,
"self": 0.000522244999956456
},
"communicator.exchange": {
"total": 0.08994305099986377,
"count": 1,
"is_parallel": true,
"self": 0.08994305099986377
},
"steps_from_proto": {
"total": 0.001869023000153902,
"count": 1,
"is_parallel": true,
"self": 0.000434095000173329,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001434927999980573,
"count": 8,
"is_parallel": true,
"self": 0.001434927999980573
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1051.7703124919306,
"count": 61700,
"is_parallel": true,
"self": 34.18507433210402,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 25.68563660493419,
"count": 61700,
"is_parallel": true,
"self": 25.68563660493419
},
"communicator.exchange": {
"total": 886.8563374479554,
"count": 61700,
"is_parallel": true,
"self": 886.8563374479554
},
"steps_from_proto": {
"total": 105.04326410693693,
"count": 61700,
"is_parallel": true,
"self": 21.13546967494017,
"children": {
"_process_rank_one_or_two_observation": {
"total": 83.90779443199676,
"count": 493600,
"is_parallel": true,
"self": 83.90779443199676
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 535.4020956799657,
"count": 61701,
"self": 2.3336598010114358,
"children": {
"process_trajectory": {
"total": 114.3539598499517,
"count": 61701,
"self": 114.11102474695122,
"children": {
"RLTrainer._checkpoint": {
"total": 0.24293510300049093,
"count": 2,
"self": 0.24293510300049093
}
}
},
"_update_policy": {
"total": 418.7144760290025,
"count": 214,
"self": 278.0704140559651,
"children": {
"TorchPPOOptimizer.update": {
"total": 140.64406197303742,
"count": 11139,
"self": 140.64406197303742
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.020999661704991e-06,
"count": 1,
"self": 1.020999661704991e-06
},
"TrainerController._save_models": {
"total": 0.10183321200020146,
"count": 1,
"self": 0.0023347570004261797,
"children": {
"RLTrainer._checkpoint": {
"total": 0.09949845499977528,
"count": 1,
"self": 0.09949845499977528
}
}
}
}
}
}
}