ppo-Pyramids / run_logs /timers.json
Kolosok's picture
First Push
cf5178d verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.6507675647735596,
"min": 0.5509337782859802,
"max": 1.4129174947738647,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 19200.24609375,
"min": 16580.90234375,
"max": 42862.265625,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989955.0,
"min": 29952.0,
"max": 989955.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989955.0,
"min": 29952.0,
"max": 989955.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.129915252327919,
"min": -0.1009550616145134,
"max": 0.129915252327919,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 32.47881317138672,
"min": -24.128259658813477,
"max": 32.47881317138672,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.02870396338403225,
"min": -0.0077840061858296394,
"max": 0.2500523030757904,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 7.175991058349609,
"min": -1.9070814847946167,
"max": 60.262603759765625,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06598723436140323,
"min": 0.06560419254329908,
"max": 0.0725121056494977,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.9238212810596452,
"min": 0.5023384671905827,
"max": 1.060396611748729,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.007367170281170167,
"min": 7.913247748531298e-05,
"max": 0.008544848188111211,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.10314038393638233,
"min": 0.0011078546847943817,
"max": 0.11962787463355697,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.345918979964283e-06,
"min": 7.345918979964283e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00010284286571949997,
"min": 0.00010284286571949997,
"max": 0.003634430588523199,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10244860714285713,
"min": 0.10244860714285713,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4342804999999998,
"min": 1.3886848,
"max": 2.6114768000000006,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0002546158535714286,
"min": 0.0002546158535714286,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0035646219499999997,
"min": 0.0035646219499999997,
"max": 0.12116653232000002,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.013214514590799809,
"min": 0.013214514590799809,
"max": 0.46827688813209534,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.18500320613384247,
"min": 0.18500320613384247,
"max": 3.2779381275177,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 756.7631578947369,
"min": 691.5238095238095,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 28757.0,
"min": 15984.0,
"max": 32728.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 0.4927837502312016,
"min": -1.0000000521540642,
"max": 0.5352097157298065,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 18.23299875855446,
"min": -30.991201616823673,
"max": 22.290598087012768,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 0.4927837502312016,
"min": -1.0000000521540642,
"max": 0.5352097157298065,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 18.23299875855446,
"min": -30.991201616823673,
"max": 22.290598087012768,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.10148470406105893,
"min": 0.09966874672229298,
"max": 9.230249600484967,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 3.7549340502591804,
"min": 3.7549340502591804,
"max": 147.68399360775948,
"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": "1764932054",
"python_version": "3.10.12 | packaged by conda-forge | (main, Jun 23 2023, 22:40:32) [GCC 12.3.0]",
"command_line_arguments": "/usr/local/envs/myenv/bin/mlagents-learn /content/ml-agents/config/ppo/PyramidsRND.yaml --env=./training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids Training --no-graphics --force",
"mlagents_version": "1.2.0.dev0",
"mlagents_envs_version": "1.2.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.8.0+cu128",
"numpy_version": "1.23.5",
"end_time_seconds": "1764934083"
},
"total": 2028.862407224,
"count": 1,
"self": 0.47641210200072237,
"children": {
"run_training.setup": {
"total": 0.02024339699983102,
"count": 1,
"self": 0.02024339699983102
},
"TrainerController.start_learning": {
"total": 2028.3657517249994,
"count": 1,
"self": 1.2985112390415452,
"children": {
"TrainerController._reset_env": {
"total": 2.4870154000000184,
"count": 1,
"self": 2.4870154000000184
},
"TrainerController.advance": {
"total": 2024.506140657958,
"count": 63283,
"self": 1.3377380561123573,
"children": {
"env_step": {
"total": 1377.7605678278624,
"count": 63283,
"self": 1233.306622854042,
"children": {
"SubprocessEnvManager._take_step": {
"total": 143.66314934085767,
"count": 63283,
"self": 4.530403729977479,
"children": {
"TorchPolicy.evaluate": {
"total": 139.1327456108802,
"count": 62554,
"self": 139.1327456108802
}
}
},
"workers": {
"total": 0.7907956329627268,
"count": 63283,
"self": 0.0,
"children": {
"worker_root": {
"total": 2022.0946130499697,
"count": 63283,
"is_parallel": true,
"self": 900.6530895229225,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0020821640000576735,
"count": 1,
"is_parallel": true,
"self": 0.0006800660021326621,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0014020979979250114,
"count": 8,
"is_parallel": true,
"self": 0.0014020979979250114
}
}
},
"UnityEnvironment.step": {
"total": 0.09546161300022504,
"count": 1,
"is_parallel": true,
"self": 0.000515534999067313,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0004152250003244262,
"count": 1,
"is_parallel": true,
"self": 0.0004152250003244262
},
"communicator.exchange": {
"total": 0.09309478900013346,
"count": 1,
"is_parallel": true,
"self": 0.09309478900013346
},
"steps_from_proto": {
"total": 0.0014360640006998437,
"count": 1,
"is_parallel": true,
"self": 0.0002959529983854736,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011401110023143701,
"count": 8,
"is_parallel": true,
"self": 0.0011401110023143701
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1121.4415235270471,
"count": 63282,
"is_parallel": true,
"self": 32.256380894042195,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 21.777582994037402,
"count": 63282,
"is_parallel": true,
"self": 21.777582994037402
},
"communicator.exchange": {
"total": 964.0986574029985,
"count": 63282,
"is_parallel": true,
"self": 964.0986574029985
},
"steps_from_proto": {
"total": 103.30890223596907,
"count": 63282,
"is_parallel": true,
"self": 22.082029344038347,
"children": {
"_process_rank_one_or_two_observation": {
"total": 81.22687289193073,
"count": 506256,
"is_parallel": true,
"self": 81.22687289193073
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 645.4078347739833,
"count": 63283,
"self": 2.5759363600063807,
"children": {
"process_trajectory": {
"total": 119.7648529069811,
"count": 63283,
"self": 119.58080011098082,
"children": {
"RLTrainer._checkpoint": {
"total": 0.18405279600028734,
"count": 2,
"self": 0.18405279600028734
}
}
},
"_update_policy": {
"total": 523.0670455069958,
"count": 451,
"self": 292.18601921597656,
"children": {
"TorchPPOOptimizer.update": {
"total": 230.88102629101923,
"count": 22830,
"self": 230.88102629101923
}
}
}
}
}
}
},
"trainer_threads": {
"total": 7.779999577905983e-07,
"count": 1,
"self": 7.779999577905983e-07
},
"TrainerController._save_models": {
"total": 0.07408364999992045,
"count": 1,
"self": 0.0012313799998082686,
"children": {
"RLTrainer._checkpoint": {
"total": 0.07285227000011218,
"count": 1,
"self": 0.07285227000011218
}
}
}
}
}
}
}