ppo-Pyramid / run_logs /timers.json
devjwsong's picture
First commit
ffdb3a0 verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.4875069856643677,
"min": 0.4875069856643677,
"max": 1.4374103546142578,
"count": 30
},
"Pyramids.Policy.Entropy.sum": {
"value": 14617.4091796875,
"min": 14617.4091796875,
"max": 43605.28125,
"count": 30
},
"Pyramids.Step.mean": {
"value": 899997.0,
"min": 29952.0,
"max": 899997.0,
"count": 30
},
"Pyramids.Step.sum": {
"value": 899997.0,
"min": 29952.0,
"max": 899997.0,
"count": 30
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.5028297305107117,
"min": -0.11015240103006363,
"max": 0.5086981654167175,
"count": 30
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 137.27252197265625,
"min": -26.546728134155273,
"max": 141.3079071044922,
"count": 30
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.01257468480616808,
"min": -0.009992149658501148,
"max": 0.587675154209137,
"count": 30
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 3.432888984680176,
"min": -2.7078726291656494,
"max": 139.2790069580078,
"count": 30
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.0684604044931586,
"min": 0.06530274381790484,
"max": 0.07273378707484934,
"count": 30
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9584456629042204,
"min": 0.5070467034059418,
"max": 1.064435244499085,
"count": 30
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.019412931133826663,
"min": 0.00021315773422659592,
"max": 0.019412931133826663,
"count": 30
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.2717810358735733,
"min": 0.002557892810719151,
"max": 0.2717810358735733,
"count": 30
},
"Pyramids.Policy.LearningRate.mean": {
"value": 3.449443135902857e-05,
"min": 3.449443135902857e-05,
"max": 0.00029515063018788575,
"count": 30
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.0004829220390264,
"min": 0.0004829220390264,
"max": 0.0035091206302931997,
"count": 30
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.1114981142857143,
"min": 0.1114981142857143,
"max": 0.19838354285714285,
"count": 30
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.5609736,
"min": 1.3886848,
"max": 2.5697067999999996,
"count": 30
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0011586616171428573,
"min": 0.0011586616171428573,
"max": 0.00983851593142857,
"count": 30
},
"Pyramids.Policy.Beta.sum": {
"value": 0.01622126264,
"min": 0.01622126264,
"max": 0.11699370932,
"count": 30
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.0116698844358325,
"min": 0.0116698844358325,
"max": 0.5449455976486206,
"count": 30
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.16337838768959045,
"min": 0.16337838768959045,
"max": 3.814619302749634,
"count": 30
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 340.25555555555553,
"min": 340.25555555555553,
"max": 999.0,
"count": 30
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 30623.0,
"min": 15984.0,
"max": 33573.0,
"count": 30
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.6152777550948991,
"min": -1.0000000521540642,
"max": 1.6152777550948991,
"count": 30
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 145.37499795854092,
"min": -30.32340171933174,
"max": 145.37499795854092,
"count": 30
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.6152777550948991,
"min": -1.0000000521540642,
"max": 1.6152777550948991,
"count": 30
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 145.37499795854092,
"min": -30.32340171933174,
"max": 145.37499795854092,
"count": 30
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.041155628341786925,
"min": 0.041155628341786925,
"max": 11.827046897262335,
"count": 30
},
"Pyramids.Policy.RndReward.sum": {
"value": 3.7040065507608233,
"min": 3.585337777272798,
"max": 189.23275035619736,
"count": 30
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 30
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 30
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1715656839",
"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/Pyramids/Pyramids --run-id=Pyramids Training --no-graphics",
"mlagents_version": "1.1.0.dev0",
"mlagents_envs_version": "1.1.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.2.1+cu121",
"numpy_version": "1.23.5",
"end_time_seconds": "1715658980"
},
"total": 2140.343920952,
"count": 1,
"self": 0.33082992200070294,
"children": {
"run_training.setup": {
"total": 0.05049393799981772,
"count": 1,
"self": 0.05049393799981772
},
"TrainerController.start_learning": {
"total": 2139.9625970919997,
"count": 1,
"self": 1.4642647269529334,
"children": {
"TrainerController._reset_env": {
"total": 2.065287731999888,
"count": 1,
"self": 2.065287731999888
},
"TrainerController.advance": {
"total": 2136.278141518047,
"count": 59224,
"self": 1.4668728161036597,
"children": {
"env_step": {
"total": 1529.8349801950694,
"count": 59224,
"self": 1398.2566768131483,
"children": {
"SubprocessEnvManager._take_step": {
"total": 130.71230611893907,
"count": 59224,
"self": 4.737067057018521,
"children": {
"TorchPolicy.evaluate": {
"total": 125.97523906192055,
"count": 58133,
"self": 125.97523906192055
}
}
},
"workers": {
"total": 0.8659972629820913,
"count": 59223,
"self": 0.0,
"children": {
"worker_root": {
"total": 2135.081124265029,
"count": 59223,
"is_parallel": true,
"self": 858.391979879028,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.002117945999998483,
"count": 1,
"is_parallel": true,
"self": 0.0006663509989266458,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001451595001071837,
"count": 8,
"is_parallel": true,
"self": 0.001451595001071837
}
}
},
"UnityEnvironment.step": {
"total": 0.06801497099968401,
"count": 1,
"is_parallel": true,
"self": 0.0006631749993175617,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0004428400002325361,
"count": 1,
"is_parallel": true,
"self": 0.0004428400002325361
},
"communicator.exchange": {
"total": 0.06423211099991022,
"count": 1,
"is_parallel": true,
"self": 0.06423211099991022
},
"steps_from_proto": {
"total": 0.002676845000223693,
"count": 1,
"is_parallel": true,
"self": 0.00041974799933086615,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.002257097000892827,
"count": 8,
"is_parallel": true,
"self": 0.002257097000892827
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1276.6891443860009,
"count": 59222,
"is_parallel": true,
"self": 32.409322104204875,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 22.579389575836103,
"count": 59222,
"is_parallel": true,
"self": 22.579389575836103
},
"communicator.exchange": {
"total": 1125.4354111580114,
"count": 59222,
"is_parallel": true,
"self": 1125.4354111580114
},
"steps_from_proto": {
"total": 96.26502154794844,
"count": 59222,
"is_parallel": true,
"self": 19.967355799552024,
"children": {
"_process_rank_one_or_two_observation": {
"total": 76.29766574839641,
"count": 473776,
"is_parallel": true,
"self": 76.29766574839641
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 604.976288506874,
"count": 59223,
"self": 2.61674405488111,
"children": {
"process_trajectory": {
"total": 126.52306561499518,
"count": 59223,
"self": 126.31990286599466,
"children": {
"RLTrainer._checkpoint": {
"total": 0.20316274900051212,
"count": 1,
"self": 0.20316274900051212
}
}
},
"_update_policy": {
"total": 475.8364788369977,
"count": 416,
"self": 279.5938668220142,
"children": {
"TorchPPOOptimizer.update": {
"total": 196.24261201498348,
"count": 21219,
"self": 196.24261201498348
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.3549997674999759e-06,
"count": 1,
"self": 1.3549997674999759e-06
},
"TrainerController._save_models": {
"total": 0.15490176000002975,
"count": 1,
"self": 0.002277440000398201,
"children": {
"RLTrainer._checkpoint": {
"total": 0.15262431999963155,
"count": 1,
"self": 0.15262431999963155
}
}
}
}
}
}
}