FrancoisDongier's picture
Pyramids first push
dfe256e
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.3476437032222748,
"min": 0.3476437032222748,
"max": 1.5062692165374756,
"count": 33
},
"Pyramids.Policy.Entropy.sum": {
"value": 10423.7490234375,
"min": 10423.7490234375,
"max": 45694.18359375,
"count": 33
},
"Pyramids.Step.mean": {
"value": 989945.0,
"min": 29952.0,
"max": 989945.0,
"count": 33
},
"Pyramids.Step.sum": {
"value": 989945.0,
"min": 29952.0,
"max": 989945.0,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.6079534292221069,
"min": -0.10254489630460739,
"max": 0.6354588270187378,
"count": 33
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 171.44287109375,
"min": -24.713319778442383,
"max": 181.10577392578125,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": -0.1059713065624237,
"min": -0.1059713065624237,
"max": 0.18817371129989624,
"count": 33
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": -29.883909225463867,
"min": -29.883909225463867,
"max": 45.16168975830078,
"count": 33
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.07094125884274642,
"min": 0.06490606364552928,
"max": 0.07313080538741992,
"count": 33
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 1.0641188826411963,
"min": 0.5007884053290278,
"max": 1.0791497594618704,
"count": 33
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.015913318131869244,
"min": 0.0007362922636559103,
"max": 0.01780703528631767,
"count": 33
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.23869977197803868,
"min": 0.007526886338204892,
"max": 0.25175333258812316,
"count": 33
},
"Pyramids.Policy.LearningRate.mean": {
"value": 7.502997499033333e-06,
"min": 7.502997499033333e-06,
"max": 0.00029515063018788575,
"count": 33
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.00011254496248549999,
"min": 0.00011254496248549999,
"max": 0.0035083598305467997,
"count": 33
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10250096666666667,
"min": 0.10250096666666667,
"max": 0.19838354285714285,
"count": 33
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.5375145,
"min": 1.3691136000000002,
"max": 2.5694532000000003,
"count": 33
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00025984657,
"min": 0.00025984657,
"max": 0.00983851593142857,
"count": 33
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0038976985499999998,
"min": 0.0038976985499999998,
"max": 0.11696837468,
"count": 33
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.010266873985528946,
"min": 0.010266873985528946,
"max": 0.2730003297328949,
"count": 33
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.1540031135082245,
"min": 0.14740155637264252,
"max": 1.911002278327942,
"count": 33
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 309.1290322580645,
"min": 303.1818181818182,
"max": 999.0,
"count": 33
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 28749.0,
"min": 15984.0,
"max": 33402.0,
"count": 33
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.6693591195088562,
"min": -1.0000000521540642,
"max": 1.6913043312404468,
"count": 33
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 155.25039811432362,
"min": -32.000001668930054,
"max": 163.9839982241392,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.6693591195088562,
"min": -1.0000000521540642,
"max": 1.6913043312404468,
"count": 33
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 155.25039811432362,
"min": -32.000001668930054,
"max": 163.9839982241392,
"count": 33
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.03264565658611646,
"min": 0.03264565658611646,
"max": 4.987187007442117,
"count": 33
},
"Pyramids.Policy.RndReward.sum": {
"value": 3.036046062508831,
"min": 3.036046062508831,
"max": 79.79499211907387,
"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": "1677281864",
"python_version": "3.8.10 (default, Nov 14 2022, 12:59:47) \n[GCC 9.4.0]",
"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": "0.29.0.dev0",
"mlagents_envs_version": "0.29.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "1.8.1+cu102",
"numpy_version": "1.22.4",
"end_time_seconds": "1677284129"
},
"total": 2264.297855807,
"count": 1,
"self": 0.4379792670001734,
"children": {
"run_training.setup": {
"total": 0.11615187600000354,
"count": 1,
"self": 0.11615187600000354
},
"TrainerController.start_learning": {
"total": 2263.7437246639997,
"count": 1,
"self": 1.2570200669420046,
"children": {
"TrainerController._reset_env": {
"total": 7.414926612000272,
"count": 1,
"self": 7.414926612000272
},
"TrainerController.advance": {
"total": 2254.9888350740575,
"count": 63929,
"self": 1.3454794750773544,
"children": {
"env_step": {
"total": 1502.6277517049284,
"count": 63929,
"self": 1393.5167879280139,
"children": {
"SubprocessEnvManager._take_step": {
"total": 108.32629039501262,
"count": 63929,
"self": 4.6155019620946405,
"children": {
"TorchPolicy.evaluate": {
"total": 103.71078843291798,
"count": 62560,
"self": 35.4735873769032,
"children": {
"TorchPolicy.sample_actions": {
"total": 68.23720105601478,
"count": 62560,
"self": 68.23720105601478
}
}
}
}
},
"workers": {
"total": 0.7846733819019391,
"count": 63929,
"self": 0.0,
"children": {
"worker_root": {
"total": 2259.0954466958997,
"count": 63929,
"is_parallel": true,
"self": 981.535894725821,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0018701059998420533,
"count": 1,
"is_parallel": true,
"self": 0.000677740999890375,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011923649999516783,
"count": 8,
"is_parallel": true,
"self": 0.0011923649999516783
}
}
},
"UnityEnvironment.step": {
"total": 0.04527602300004219,
"count": 1,
"is_parallel": true,
"self": 0.0005296640001688502,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.00046416500026680296,
"count": 1,
"is_parallel": true,
"self": 0.00046416500026680296
},
"communicator.exchange": {
"total": 0.04276678599990191,
"count": 1,
"is_parallel": true,
"self": 0.04276678599990191
},
"steps_from_proto": {
"total": 0.001515407999704621,
"count": 1,
"is_parallel": true,
"self": 0.0003761319999284751,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.001139275999776146,
"count": 8,
"is_parallel": true,
"self": 0.001139275999776146
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1277.5595519700787,
"count": 63928,
"is_parallel": true,
"self": 30.67958531810791,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 22.570543732003898,
"count": 63928,
"is_parallel": true,
"self": 22.570543732003898
},
"communicator.exchange": {
"total": 1136.085815266004,
"count": 63928,
"is_parallel": true,
"self": 1136.085815266004
},
"steps_from_proto": {
"total": 88.22360765396297,
"count": 63928,
"is_parallel": true,
"self": 20.64059371077019,
"children": {
"_process_rank_one_or_two_observation": {
"total": 67.58301394319278,
"count": 511424,
"is_parallel": true,
"self": 67.58301394319278
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 751.0156038940518,
"count": 63929,
"self": 2.4404530590404647,
"children": {
"process_trajectory": {
"total": 165.86683003101416,
"count": 63929,
"self": 165.68649488601386,
"children": {
"RLTrainer._checkpoint": {
"total": 0.18033514500029924,
"count": 2,
"self": 0.18033514500029924
}
}
},
"_update_policy": {
"total": 582.7083208039971,
"count": 450,
"self": 229.96572979894518,
"children": {
"TorchPPOOptimizer.update": {
"total": 352.74259100505196,
"count": 22761,
"self": 352.74259100505196
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.2299997251830064e-06,
"count": 1,
"self": 1.2299997251830064e-06
},
"TrainerController._save_models": {
"total": 0.08294168100019306,
"count": 1,
"self": 0.0015078970000104164,
"children": {
"RLTrainer._checkpoint": {
"total": 0.08143378400018264,
"count": 1,
"self": 0.08143378400018264
}
}
}
}
}
}
}