akanametov's picture
Trained Pyramids agent upload
346cc4f
{
"name": "root",
"gauges": {
"Pyramids.Policy.Entropy.mean": {
"value": 0.2673548758029938,
"min": 0.263279527425766,
"max": 1.5107438564300537,
"count": 50
},
"Pyramids.Policy.Entropy.sum": {
"value": 7982.14697265625,
"min": 7974.2099609375,
"max": 45829.92578125,
"count": 50
},
"Pyramids.Step.mean": {
"value": 1499954.0,
"min": 29952.0,
"max": 1499954.0,
"count": 50
},
"Pyramids.Step.sum": {
"value": 1499954.0,
"min": 29952.0,
"max": 1499954.0,
"count": 50
},
"Pyramids.Policy.ExtrinsicValueEstimate.mean": {
"value": 0.5911348462104797,
"min": -0.09513404965400696,
"max": 0.7207068800926208,
"count": 50
},
"Pyramids.Policy.ExtrinsicValueEstimate.sum": {
"value": 166.7000274658203,
"min": -22.832172393798828,
"max": 209.7257080078125,
"count": 50
},
"Pyramids.Policy.RndValueEstimate.mean": {
"value": 0.008362549357116222,
"min": -0.016310542821884155,
"max": 0.18181274831295013,
"count": 50
},
"Pyramids.Policy.RndValueEstimate.sum": {
"value": 2.358238935470581,
"min": -4.697436332702637,
"max": 43.816871643066406,
"count": 50
},
"Pyramids.Losses.PolicyLoss.mean": {
"value": 0.07153249748877162,
"min": 0.06373093732271058,
"max": 0.07552220193191553,
"count": 50
},
"Pyramids.Losses.PolicyLoss.sum": {
"value": 1.0014549648428026,
"min": 0.4871466045603076,
"max": 1.055715087361807,
"count": 50
},
"Pyramids.Losses.ValueLoss.mean": {
"value": 0.01677720902426656,
"min": 0.00013691623318073045,
"max": 0.01772582845210092,
"count": 50
},
"Pyramids.Losses.ValueLoss.sum": {
"value": 0.23488092633973184,
"min": 0.001506078564988035,
"max": 0.2529735965882519,
"count": 50
},
"Pyramids.Policy.LearningRate.mean": {
"value": 2.886999037699996e-06,
"min": 2.886999037699996e-06,
"max": 0.00029676708679192377,
"count": 50
},
"Pyramids.Policy.LearningRate.sum": {
"value": 4.0417986527799945e-05,
"min": 4.0417986527799945e-05,
"max": 0.003654744581751866,
"count": 50
},
"Pyramids.Policy.Epsilon.mean": {
"value": 0.10096230000000002,
"min": 0.10096230000000002,
"max": 0.19892236190476195,
"count": 50
},
"Pyramids.Policy.Epsilon.sum": {
"value": 1.4134722000000002,
"min": 1.3924565333333336,
"max": 2.618248133333333,
"count": 50
},
"Pyramids.Policy.Beta.mean": {
"value": 0.00010613376999999983,
"min": 0.00010613376999999983,
"max": 0.009892343954285714,
"count": 50
},
"Pyramids.Policy.Beta.sum": {
"value": 0.0014858727799999976,
"min": 0.0014858727799999976,
"max": 0.12184298852,
"count": 50
},
"Pyramids.Losses.RNDLoss.mean": {
"value": 0.0067257629707455635,
"min": 0.0067257629707455635,
"max": 0.285749614238739,
"count": 50
},
"Pyramids.Losses.RNDLoss.sum": {
"value": 0.09416068345308304,
"min": 0.09416068345308304,
"max": 2.0002472400665283,
"count": 50
},
"Pyramids.Environment.EpisodeLength.mean": {
"value": 292.08,
"min": 261.38260869565215,
"max": 999.0,
"count": 50
},
"Pyramids.Environment.EpisodeLength.sum": {
"value": 29208.0,
"min": 15984.0,
"max": 32427.0,
"count": 50
},
"Pyramids.Environment.CumulativeReward.mean": {
"value": 1.6879139818251132,
"min": -1.0000000521540642,
"max": 1.7286666479927522,
"count": 50
},
"Pyramids.Environment.CumulativeReward.sum": {
"value": 168.79139818251133,
"min": -31.99760167300701,
"max": 197.94019880890846,
"count": 50
},
"Pyramids.Policy.ExtrinsicReward.mean": {
"value": 1.6879139818251132,
"min": -1.0000000521540642,
"max": 1.7286666479927522,
"count": 50
},
"Pyramids.Policy.ExtrinsicReward.sum": {
"value": 168.79139818251133,
"min": -31.99760167300701,
"max": 197.94019880890846,
"count": 50
},
"Pyramids.Policy.RndReward.mean": {
"value": 0.020380923647608143,
"min": 0.019019794243333468,
"max": 5.5465923440642655,
"count": 50
},
"Pyramids.Policy.RndReward.sum": {
"value": 2.0380923647608142,
"min": 2.0380923647608142,
"max": 88.74547750502825,
"count": 50
},
"Pyramids.IsTraining.mean": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 50
},
"Pyramids.IsTraining.sum": {
"value": 1.0,
"min": 1.0,
"max": 1.0,
"count": 50
}
},
"metadata": {
"timer_format_version": "0.1.0",
"start_time_seconds": "1671961511",
"python_version": "3.8.16 (default, Dec 7 2022, 01:12:13) \n[GCC 7.5.0]",
"command_line_arguments": "/usr/local/bin/mlagents-learn ./config/ppo/PyramidsRND.yaml --env=./trained-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": "1671964669"
},
"total": 3158.701565877,
"count": 1,
"self": 0.4394406699998399,
"children": {
"run_training.setup": {
"total": 0.11346505000005891,
"count": 1,
"self": 0.11346505000005891
},
"TrainerController.start_learning": {
"total": 3158.148660157,
"count": 1,
"self": 1.7506262630467972,
"children": {
"TrainerController._reset_env": {
"total": 9.678054705000022,
"count": 1,
"self": 9.678054705000022
},
"TrainerController.advance": {
"total": 3146.6286121919534,
"count": 96375,
"self": 1.8971350499250548,
"children": {
"env_step": {
"total": 2160.5217673660286,
"count": 96375,
"self": 2007.719304839075,
"children": {
"SubprocessEnvManager._take_step": {
"total": 151.65298578894442,
"count": 96375,
"self": 6.440625716936097,
"children": {
"TorchPolicy.evaluate": {
"total": 145.21236007200832,
"count": 93799,
"self": 49.4434342530227,
"children": {
"TorchPolicy.sample_actions": {
"total": 95.76892581898562,
"count": 93799,
"self": 95.76892581898562
}
}
}
}
},
"workers": {
"total": 1.149476738009298,
"count": 96375,
"self": 0.0,
"children": {
"worker_root": {
"total": 3151.986898460081,
"count": 96375,
"is_parallel": true,
"self": 1293.873586271121,
"children": {
"run_training.setup": {
"total": 0.0,
"count": 0,
"is_parallel": true,
"self": 0.0,
"children": {
"steps_from_proto": {
"total": 0.006428441000025487,
"count": 1,
"is_parallel": true,
"self": 0.0038612920001241946,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.002567148999901292,
"count": 8,
"is_parallel": true,
"self": 0.002567148999901292
}
}
},
"UnityEnvironment.step": {
"total": 0.049320203999968726,
"count": 1,
"is_parallel": true,
"self": 0.0005554960000608844,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 0.0005353560000003199,
"count": 1,
"is_parallel": true,
"self": 0.0005353560000003199
},
"communicator.exchange": {
"total": 0.04637663500000144,
"count": 1,
"is_parallel": true,
"self": 0.04637663500000144
},
"steps_from_proto": {
"total": 0.0018527169999060789,
"count": 1,
"is_parallel": true,
"self": 0.0005125980001139396,
"children": {
"_process_rank_one_or_two_observation": {
"total": 0.0013401189997921392,
"count": 8,
"is_parallel": true,
"self": 0.0013401189997921392
}
}
}
}
}
}
},
"UnityEnvironment.step": {
"total": 1858.1133121889602,
"count": 96374,
"is_parallel": true,
"self": 41.47324370997467,
"children": {
"UnityEnvironment._generate_step_input": {
"total": 34.54540494997502,
"count": 96374,
"is_parallel": true,
"self": 34.54540494997502
},
"communicator.exchange": {
"total": 1639.7156951339853,
"count": 96374,
"is_parallel": true,
"self": 1639.7156951339853
},
"steps_from_proto": {
"total": 142.37896839502548,
"count": 96374,
"is_parallel": true,
"self": 33.65286569586215,
"children": {
"_process_rank_one_or_two_observation": {
"total": 108.72610269916333,
"count": 770992,
"is_parallel": true,
"self": 108.72610269916333
}
}
}
}
}
}
}
}
}
}
},
"trainer_advance": {
"total": 984.2097097759998,
"count": 96375,
"self": 3.4083143249507657,
"children": {
"process_trajectory": {
"total": 223.10359415504672,
"count": 96375,
"self": 222.818370197047,
"children": {
"RLTrainer._checkpoint": {
"total": 0.2852239579997331,
"count": 3,
"self": 0.2852239579997331
}
}
},
"_update_policy": {
"total": 757.6978012960024,
"count": 679,
"self": 294.6853354990152,
"children": {
"TorchPPOOptimizer.update": {
"total": 463.01246579698716,
"count": 34245,
"self": 463.01246579698716
}
}
}
}
}
}
},
"trainer_threads": {
"total": 1.0630001270328648e-06,
"count": 1,
"self": 1.0630001270328648e-06
},
"TrainerController._save_models": {
"total": 0.09136593399989579,
"count": 1,
"self": 0.0013101469999128312,
"children": {
"RLTrainer._checkpoint": {
"total": 0.09005578699998296,
"count": 1,
"self": 0.09005578699998296
}
}
}
}
}
}
}