ppo-Pyramids / run_logs /timers.json
FelipePasquevich's picture
First Push
488a09a
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.8817382454872131,
"min": 0.8683117628097534,
"max": 1.5948829650878906,
"count": 10
},
"Pyramids.Policy.Entropy.sum": {
"value": 8789.1669921875,
"min": 8382.5791015625,
"max": 16331.6015625,
"count": 10
},
"Pyramids.Step.mean": {
"value": 99932.0,
"min": 9984.0,
"max": 99932.0,
"count": 10
},
"Pyramids.Step.sum": {
"value": 99932.0,
"min": 9984.0,
"max": 99932.0,
"count": 10
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": -0.09606429934501648,
"min": -0.11559466272592545,
"max": -0.04426305368542671,
"count": 10
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": -7.781208515167236,
"min": -9.247572898864746,
"max": -3.4525182247161865,
"count": 10
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.13280153274536133,
"min": 0.07141053676605225,
"max": 0.44847846031188965,
"count": 10
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 10.75692367553711,
"min": 5.570022106170654,
"max": 34.981319427490234,
"count": 10
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.07297117875115124,
"min": 0.06841447311191227,
"max": 0.07474737928229538,
"count": 10
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.29188471500460494,
"min": 0.14673426639269932,
"max": 0.36393850943635164,
"count": 10
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.001996506074277147,
"min": 0.00045420435570818366,
"max": 0.018218710143022086,
"count": 10
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.007986024297108587,
"min": 0.0018168174228327346,
"max": 0.03643742028604417,
"count": 10
},
"Pyramids.Policy.LearningRate.mean": {
"value": 1.4361095212999989e-05,
"min": 1.4361095212999989e-05,
"max": 0.00028156800614399997,
"count": 10
},
"Pyramids.Policy.LearningRate.sum": {
"value": 5.7444380851999956e-05,
"min": 5.7444380851999956e-05,
"max": 0.000826311224563,
"count": 10
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10478699999999999,
"min": 0.10478699999999999,
"max": 0.19385600000000003,
"count": 10
},
"Pyramids.Policy.Epsilon.sum": {
"value": 0.41914799999999997,
"min": 0.3319039999999999,
"max": 0.7754370000000002,
"count": 10
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00048822129999999983,
"min": 0.00048822129999999983,
"max": 0.0093862144,
"count": 10
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0019528851999999993,
"min": 0.0019528851999999993,
"max": 0.0275661563,
"count": 10
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.10996373742818832,
"min": 0.10996373742818832,
"max": 1.0417176485061646,
"count": 10
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.4398549497127533,
"min": 0.4398549497127533,
"max": 2.083435297012329,
"count": 10
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 10
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 10
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 968.2142857142857,
"min": 693.6666666666666,
"max": 999.0,
"count": 8
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 13555.0,
"min": 2081.0,
"max": 15984.0,
"count": 8
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": -0.826200050434896,
"min": -1.0000000521540642,
"max": 0.639399935801824,
"count": 8
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": -11.566800706088543,
"min": -16.000000834465027,
"max": 1.9181998074054718,
"count": 8
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": -0.826200050434896,
"min": -1.0000000521540642,
"max": 0.639399935801824,
"count": 8
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": -11.566800706088543,
"min": -16.000000834465027,
"max": 1.9181998074054718,
"count": 8
},
"Pyramids.Policy.RndReward.mean": {
"value": 1.1532861574419908,
"min": 0.9339636117219925,
"max": 11.449633229523897,
"count": 8
},
"Pyramids.Policy.RndReward.sum": {
"value": 16.14600620418787,
"min": 2.8018908351659775,
"max": 183.19413167238235,
"count": 8
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1681940624",
"python_version": "3.9.16 (main, Dec 7 2022, 01:11:51) \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": "0.31.0.dev0",
"mlagents_envs_version": "0.31.0.dev0",
"communication_protocol_version": "1.5.0",
"pytorch_version": "1.11.0+cu102",
"numpy_version": "1.21.2",
"end_time_seconds": "1681940816"
},
"total": 192.4735251259999,
"count": 1,
"self": 0.4355843149996872,
"children": {
"run_training.setup": {
"total": 0.10713875100009318,
"count": 1,
"self": 0.10713875100009318
},
"TrainerController.start_learning": {
"total": 191.93080206000013,
"count": 1,
"self": 0.11634709298778034,
"children": {
"TrainerController._reset_env": {
"total": 4.882134713999221,
"count": 1,
"self": 4.882134713999221
},
"TrainerController.advance": {
"total": 186.81987808201302,
"count": 6315,
"self": 0.1259490399997958,
"children": {
"env_step": {
"total": 130.12431173100504,
"count": 6315,
"self": 120.12671145799868,
"children": {
"SubprocessEnvManager._take_step": {
"total": 9.920459391024451,
"count": 6315,
"self": 0.4447257600550074,
"children": {
"TorchPolicy.evaluate": {
"total": 9.475733630969444,
"count": 6295,
"self": 9.475733630969444
}
}
},
"workers": {
"total": 0.07714088198190439,
"count": 6315,
"self": 0.0,
"children": {
"worker_root": {
"total": 191.4687631180268,
"count": 6315,
"is_parallel": true,
"self": 81.65669950604206,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0017624709998926846,
"count": 1,
"is_parallel": true,
"self": 0.0005493830003615585,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0012130879995311261,
"count": 8,
"is_parallel": true,
"self": 0.0012130879995311261
}
}
},
"UnityEnvironment.step": {
"total": 0.0763542230006351,
"count": 1,
"is_parallel": true,
"self": 0.0005415830000856658,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0005433619999166694,
"count": 1,
"is_parallel": true,
"self": 0.0005433619999166694
},
"communicator.exchange": {
"total": 0.07346515600056591,
"count": 1,
"is_parallel": true,
"self": 0.07346515600056591
},
"steps_from_proto": {
"total": 0.0018041220000668545,
"count": 1,
"is_parallel": true,
"self": 0.00043513199852895923,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0013689900015378953,
"count": 8,
"is_parallel": true,
"self": 0.0013689900015378953
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 109.81206361198474,
"count": 6314,
"is_parallel": true,
"self": 3.152400111897805,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 2.236652168055116,
"count": 6314,
"is_parallel": true,
"self": 2.236652168055116
},
"communicator.exchange": {
"total": 95.34944289701434,
"count": 6314,
"is_parallel": true,
"self": 95.34944289701434
},
"steps_from_proto": {
"total": 9.073568435017478,
"count": 6314,
"is_parallel": true,
"self": 1.863818704003279,
"children": {
"_process_rank_one_or_two_observation": {
"total": 7.209749731014199,
"count": 50512,
"is_parallel": true,
"self": 7.209749731014199
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 56.56961731100819,
"count": 6315,
"self": 0.16876614300781512,
"children": {
"process_trajectory": {
"total": 9.500082053999904,
"count": 6315,
"self": 9.500082053999904
},
"_update_policy": {
"total": 46.90076911400047,
"count": 35,
"self": 29.859427992006204,
"children": {
"TorchPPOOptimizer.update": {
"total": 17.041341121994265,
"count": 2277,
"self": 17.041341121994265
}
}
}
}
}
}
},
"trainer_threads": {
"total": 9.039995347848162e-07,
"count": 1,
"self": 9.039995347848162e-07
},
"TrainerController._save_models": {
"total": 0.11244126700057677,
"count": 1,
"self": 0.0013327160004337202,
"children": {
"RLTrainer._checkpoint": {
"total": 0.11110855100014305,
"count": 1,
"self": 0.11110855100014305
}
}
}
}
}
}
}