ppo-Pyramids1 / run_logs /timers.json
frangiral's picture
First model
b83eeff
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.9297638535499573,
"min": 0.9297638535499573,
"max": 1.554758071899414,
"count": 16
},
"Pyramids.Policy.Entropy.sum": {
"value": 27803.658203125,
"min": 27803.658203125,
"max": 47165.140625,
"count": 16
},
"Pyramids.Step.mean": {
"value": 479997.0,
"min": 29981.0,
"max": 479997.0,
"count": 16
},
"Pyramids.Step.sum": {
"value": 479997.0,
"min": 29981.0,
"max": 479997.0,
"count": 16
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.3730928897857666,
"min": -0.15198108553886414,
"max": 0.3730928897857666,
"count": 16
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 97.004150390625,
"min": -36.1714973449707,
"max": 97.004150390625,
"count": 16
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.03897610306739807,
"min": 0.022027648985385895,
"max": 0.12327360361814499,
"count": 16
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 10.133787155151367,
"min": 5.418801784515381,
"max": 29.585664749145508,
"count": 16
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.10293516181985406,
"min": 0.09458253070487006,
"max": 0.10293516181985406,
"count": 16
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 0.7205461327389784,
"min": 0.5011188115593675,
"max": 0.8106914077250729,
"count": 16
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.011013167805673513,
"min": 0.0002967954990626783,
"max": 0.011013167805673513,
"count": 16
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.07709217463971459,
"min": 0.00178077299437607,
"max": 0.07716441016238884,
"count": 16
},
"Pyramids.Policy.LearningRate.mean": {
"value": 0.0001604089751017857,
"min": 0.0001604089751017857,
"max": 0.00029466066177977993,
"count": 16
},
"Pyramids.Policy.LearningRate.sum": {
"value": 0.0011228628257125,
"min": 0.0011228628257125,
"max": 0.0019432188522603999,
"count": 16
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.15346964285714287,
"min": 0.15346964285714287,
"max": 0.19822022,
"count": 16
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.0742875,
"min": 0.9911011,
"max": 1.395994,
"count": 16
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00535161732142857,
"min": 0.00535161732142857,
"max": 0.009822199978,
"count": 16
},
"Pyramids.Policy.Beta.sum": {
"value": 0.03746132124999999,
"min": 0.03746132124999999,
"max": 0.06477918603999999,
"count": 16
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.013234114274382591,
"min": 0.013234114274382591,
"max": 0.21255290508270264,
"count": 16
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.09263879805803299,
"min": 0.09263879805803299,
"max": 1.0627645254135132,
"count": 16
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 547.9122807017544,
"min": 546.7169811320755,
"max": 999.0,
"count": 16
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 31231.0,
"min": 16780.0,
"max": 33237.0,
"count": 16
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.0660069848884617,
"min": -0.9998500514775515,
"max": 1.1135660097846445,
"count": 16
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 60.76239813864231,
"min": -31.995201647281647,
"max": 60.76239813864231,
"count": 16
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.0660069848884617,
"min": -0.9998500514775515,
"max": 1.1135660097846445,
"count": 16
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 60.76239813864231,
"min": -31.995201647281647,
"max": 60.76239813864231,
"count": 16
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.07554679786941658,
"min": 0.07554679786941658,
"max": 7.019986360388644,
"count": 16
},
"Pyramids.Policy.RndReward.sum": {
"value": 4.306167478556745,
"min": 4.2760298030334525,
"max": 119.33976812660694,
"count": 16
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 16
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 16
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1676383656",
"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": "1.21.6",
"end_time_seconds": "1676384934"
},
"total": 1278.0996835049998,
"count": 1,
"self": 0.318129186999613,
"children": {
"run_training.setup": {
"total": 0.11374618700006067,
"count": 1,
"self": 0.11374618700006067
},
"TrainerController.start_learning": {
"total": 1277.6678081310001,
"count": 1,
"self": 0.7985351539953172,
"children": {
"TrainerController._reset_env": {
"total": 6.546596541000099,
"count": 1,
"self": 6.546596541000099
},
"TrainerController.advance": {
"total": 1270.1716131250046,
"count": 32154,
"self": 0.7635053269609671,
"children": {
"env_step": {
"total": 755.239568163008,
"count": 32154,
"self": 693.5837074339634,
"children": {
"SubprocessEnvManager._take_step": {
"total": 61.17571985800646,
"count": 32154,
"self": 2.425299467998684,
"children": {
"TorchPolicy.evaluate": {
"total": 58.750420390007775,
"count": 31739,
"self": 19.54632017496965,
"children": {
"TorchPolicy.sample_actions": {
"total": 39.204100215038125,
"count": 31739,
"self": 39.204100215038125
}
}
}
}
},
"workers": {
"total": 0.480140871038202,
"count": 32154,
"self": 0.0,
"children": {
"worker_root": {
"total": 1273.4918120860038,
"count": 32154,
"is_parallel": true,
"self": 642.4081456060517,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.0019095679999736603,
"count": 1,
"is_parallel": true,
"self": 0.0007226180000543536,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011869499999193067,
"count": 8,
"is_parallel": true,
"self": 0.0011869499999193067
}
}
},
"UnityEnvironment.step": {
"total": 0.049325189000001046,
"count": 1,
"is_parallel": true,
"self": 0.0005521780001345178,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0005039719999331282,
"count": 1,
"is_parallel": true,
"self": 0.0005039719999331282
},
"communicator.exchange": {
"total": 0.04671118300007038,
"count": 1,
"is_parallel": true,
"self": 0.04671118300007038
},
"steps_from_proto": {
"total": 0.0015578559998630226,
"count": 1,
"is_parallel": true,
"self": 0.00037782599997626676,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0011800299998867558,
"count": 8,
"is_parallel": true,
"self": 0.0011800299998867558
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 631.0836664799522,
"count": 32153,
"is_parallel": true,
"self": 16.555622210930323,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 12.037043200987227,
"count": 32153,
"is_parallel": true,
"self": 12.037043200987227
},
"communicator.exchange": {
"total": 549.5780965379852,
"count": 32153,
"is_parallel": true,
"self": 549.5780965379852
},
"steps_from_proto": {
"total": 52.912904530049445,
"count": 32153,
"is_parallel": true,
"self": 11.923288244019886,
"children": {
"_process_rank_one_or_two_observation": {
"total": 40.98961628602956,
"count": 257224,
"is_parallel": true,
"self": 40.98961628602956
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 514.1685396350356,
"count": 32154,
"self": 1.3734572870584998,
"children": {
"process_trajectory": {
"total": 84.15249058197423,
"count": 32154,
"self": 84.03766815197423,
"children": {
"RLTrainer._checkpoint": {
"total": 0.1148224300000038,
"count": 1,
"self": 0.1148224300000038
}
}
},
"_update_policy": {
"total": 428.64259176600285,
"count": 117,
"self": 119.56485303299223,
"children": {
"TorchPPOOptimizer.update": {
"total": 309.0777387330106,
"count": 23498,
"self": 309.0777387330106
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.469999915570952e-06,
"count": 1,
"self": 1.469999915570952e-06
},
"TrainerController._save_models": {
"total": 0.15106184100022801,
"count": 1,
"self": 0.0019139980004183599,
"children": {
"RLTrainer._checkpoint": {
"total": 0.14914784299980965,
"count": 1,
"self": 0.14914784299980965
}
}
}
}
}
}
}