ppo-Pyramids / run_logs /timers.json
qdzzzxc's picture
First Push
895c55d verified
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.4738425314426422,
"min": 0.4738425314426422,
"max": 1.4561707973480225,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 14222.857421875,
"min": 14222.857421875,
"max": 44174.3984375,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989960.0,
"min": 29952.0,
"max": 989960.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989960.0,
"min": 29952.0,
"max": 989960.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.5484902262687683,
"min": -0.10808103531599045,
"max": 0.5822015404701233,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 150.28631591796875,
"min": -26.155611038208008,
"max": 160.6876220703125,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.008397314697504044,
"min": -0.08430320769548416,
"max": 0.25763002038002014,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 2.3008642196655273,
"min": -21.328712463378906,
"max": 61.831207275390625,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.06870939726888024,
"min": 0.0656433479923361,
"max": 0.07638829070430521,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 1.0306409590332037,
"min": 0.5054512103372548,
"max": 1.0562183156720242,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.015538455484194523,
"min": 0.00034305114307330145,
"max": 0.015538455484194523,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.23307683226291784,
"min": 0.004116613716879617,
"max": 0.23307683226291784,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.48873750378667e-06,
"min": 7.48873750378667e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00011233106255680005,
"min": 0.00011233106255680005,
"max": 0.0035073032308989994,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10249621333333335,
"min": 0.10249621333333335,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.5374432000000002,
"min": 1.3691136000000002,
"max": 2.569101,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.0002593717120000002,
"min": 0.0002593717120000002,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0038905756800000028,
"min": 0.0038905756800000028,
"max": 0.11693318990000001,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.010883810929954052,
"min": 0.010883810929954052,
"max": 0.28096893429756165,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.1632571667432785,
"min": 0.15435296297073364,
"max": 1.966782569885254,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 335.42528735632186,
"min": 335.42528735632186,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 29182.0,
"min": 15984.0,
"max": 33376.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.5934767157359178,
"min": -1.0000000521540642,
"max": 1.6268683971935196,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 137.03899755328894,
"min": -32.000001668930054,
"max": 137.9437980800867,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.5934767157359178,
"min": -1.0000000521540642,
"max": 1.6268683971935196,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 137.03899755328894,
"min": -32.000001668930054,
"max": 137.9437980800867,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.037839388307644224,
"min": 0.037839388307644224,
"max": 5.216815858148038,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 3.2541873944574036,
"min": 3.2541873944574036,
"max": 83.46905373036861,
"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": "1767900502",
"python_version": "3.10.19 (main, Dec 9 2025, 19:03:08) [Clang 21.1.4 ]",
"command_line_arguments": "/home/nikita/projects/deep_rl_course/.venv/bin/mlagents-learn /home/nikita/projects/deep_rl_course/notebooks/5_unity/PyramidsRND.yaml --env=/home/nikita/training-envs-executables/linux/Pyramids/Pyramids --run-id=Pyramids Training --no-graphics",
"mlagents_version": "1.1.0",
"mlagents_envs_version": "1.1.0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "2.1.1+cu121",
"numpy_version": "1.23.5",
"end_time_seconds": "1767901149"
},
"total": 647.098403344,
"count": 1,
"self": 0.2173804409994773,
"children": {
"run_training.setup": {
"total": 0.01034513200102083,
"count": 1,
"self": 0.01034513200102083
},
"TrainerController.start_learning": {
"total": 646.8706777709995,
"count": 1,
"self": 0.5410833849055052,
"children": {
"TrainerController._reset_env": {
"total": 1.107174037999357,
"count": 1,
"self": 1.107174037999357
},
"TrainerController.advance": {
"total": 645.1611242130966,
"count": 63777,
"self": 0.5293257022058242,
"children": {
"env_step": {
"total": 395.77770801497354,
"count": 63777,
"self": 340.05522811078663,
"children": {
"SubprocessEnvManager._take_step": {
"total": 55.35414857914293,
"count": 63777,
"self": 1.689473555992663,
"children": {
"TorchPolicy.evaluate": {
"total": 53.66467502315027,
"count": 62568,
"self": 53.66467502315027
}
}
},
"workers": {
"total": 0.36833132504398236,
"count": 63777,
"self": 0.0,
"children": {
"worker_root": {
"total": 646.0345052020002,
"count": 63777,
"is_parallel": true,
"self": 343.6075374910306,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0009784969988686498,
"count": 1,
"is_parallel": true,
"self": 0.00031062900234246626,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0006678679965261836,
"count": 8,
"is_parallel": true,
"self": 0.0006678679965261836
}
}
},
"UnityEnvironment.step": {
"total": 0.01664598599927558,
"count": 1,
"is_parallel": true,
"self": 0.00020234899966453668,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00016550899999856483,
"count": 1,
"is_parallel": true,
"self": 0.00016550899999856483
},
"communicator.exchange": {
"total": 0.01566110900057538,
"count": 1,
"is_parallel": true,
"self": 0.01566110900057538
},
"steps_from_proto": {
"total": 0.0006170189990371,
"count": 1,
"is_parallel": true,
"self": 0.0001415999959135661,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0004754190031235339,
"count": 8,
"is_parallel": true,
"self": 0.0004754190031235339
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 302.4269677109696,
"count": 63776,
"is_parallel": true,
"self": 8.46849374012163,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 5.960097347906412,
"count": 63776,
"is_parallel": true,
"self": 5.960097347906412
},
"communicator.exchange": {
"total": 264.25767465309036,
"count": 63776,
"is_parallel": true,
"self": 264.25767465309036
},
"steps_from_proto": {
"total": 23.740701969851216,
"count": 63776,
"is_parallel": true,
"self": 5.202577599711731,
"children": {
"_process_rank_one_or_two_observation": {
"total": 18.538124370139485,
"count": 510208,
"is_parallel": true,
"self": 18.538124370139485
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 248.85409049591726,
"count": 63777,
"self": 1.06868713800759,
"children": {
"process_trajectory": {
"total": 48.99206459891866,
"count": 63777,
"self": 48.895949819918314,
"children": {
"RLTrainer._checkpoint": {
"total": 0.09611477900034515,
"count": 2,
"self": 0.09611477900034515
}
}
},
"_update_policy": {
"total": 198.793338758991,
"count": 445,
"self": 112.96518414291495,
"children": {
"TorchPPOOptimizer.update": {
"total": 85.82815461607606,
"count": 22791,
"self": 85.82815461607606
}
}
}
}
}
}
},
"trainer_threads": {
"total": 5.499987310031429e-07,
"count": 1,
"self": 5.499987310031429e-07
},
"TrainerController._save_models": {
"total": 0.06129558499924315,
"count": 1,
"self": 0.0008522169991920236,
"children": {
"RLTrainer._checkpoint": {
"total": 0.06044336800005112,
"count": 1,
"self": 0.06044336800005112
}
}
}
}
}
}
}